acta energetica power engineering qtrly 03/2010

nergeticaact Electrical Power Engineering Quarterly

03/2010 number 5/year 2

ENERGA SA

Patronage

ENERGA SA

www.actaenergetica.org

featuring

MODERN REACTIVE POWER AND HIGHER HARMONIC COMPENSATION THROUGH THE UTILISATION OF STATCOM AND EFA DYNAMIC COMPENSATORSBogdan Bałkowski

POTENTIAL OF THE PROSUMER MARKET DEVELOPMENT IN THE CONTEXT OF POLISH SYSTEM CONDITIONSDamian Gadzialski

METHOD OF VOLTAGE SITUATION ASSESSMENT IN THE TRANSMISSION GRID TAKING INTO ACCOUNT THE REGULATION TECHNOLOGY OPERATIONJacek Jemielity Ksawery Opala

INVESTIGATION OF BOILING HEAT TRANSFER IN AN ELECTRICALLY HEATED TUBE BUNDLEKrzysztof KrasowskiJanusz T. Cieśliński

INCREMENTAL COST METHOD IN COST-EFFICIENCY EVALUATION OF INVESTMENT IN DISTRIBUTED GENERATIONZdzisław Kusto

MAXIMISATION VS. OPTIMISATION OF RADIAL POWER DISTRIBUTIONZbigniew Lubośny

MODERN NUCLEAR POWER TECHNOLOGIESJacek Marecki

VAWT – A KEY FOR DISTRIBUTED GENERATIONKrzysztof Żmijewski

POWER GRID DEVELOPMENT IN POLAND IN THE CONTEXT OF EU CLIMATE AND ENERGY PACKAGEKrzysztof Żmijewski Maciej M. Sokołowski

4

12

18

30

42

54

72

80

86

nergeticaact

Development of the generation sector in power supply systems is accompanied by a kind of argument between the supporters of the technologies which we may call classic, and advocates of so-called renewable energy sources. Therefore on one hand we have coal, gas, oil and nuclear technologies, and on the other, hydro-electricity, wind power, biogas, solar energy, biomass, geothermy, etc.

Rather perversely we could say that the issue has actually already been resolved – renewable technologies are going to win when we run out of energy deposits for classic sources. Or maybe even the renewables have al-ready won – after all it may be said that all energy sources are renewable, only the time to renew them varies.

The idea of renewability means that there is a source with a huge energy deposit – when compared to civilization’s needs – and it is available for use. The Sun can be considered such a source for Earth. We could say that it is the Sun – which, combined with the Earth’s movements – is responsible for winds, water circulation, the growth of plants and animals (as progenitors of coal, gas and oils) etc. Acknowledging its significant effecton the creation of the Solar System we need to say that it is also responsible for tides (result of Moon circula-tion), the existence of a liquid core of our planet and the existence and distribution of fossil fuel deposits. Finally, energy can be drawn directly from solar radiation.

On the other hand, however, we could say that from the human point of view, the time of renewability for an energy deposit which is longer than a typical lifespan (or possibly generation, year or even day) is too long to consider it renewable. According to this view, the influence of the Sun on some of the processes listed aboveshould be considered unsatisfactorily slow.

Getting back from the philosophical discussion to contemporary technology and relatively short time pe-riods – comparable with single generations of human beings – this edition of Acta Energetica presents articles related to power system development in generation carried out here and now, meaning the direct future and utilization of both technology groups mentioned above.

Development of power grids is not as polarized as in the generation branch. Evolution towards smart grids seems indisputable. Articles presented in this edition of AE discuss issues of making better use of grids, influencing energy quality and – something which is always an important issue – reliability of system operation(meaning here voltage stability).

Prof. dr hab. inż. Zbigniew Lubośny Editor-in-Chief Acta Energetica

4

Authors / Biographies

Bogdan BałkowskiPruszcz Gdański / Poland

For 10 years he has been the Vice-President of C&T Elmech based in Pruszcz Gdański. He manages the R&D Department operation and sets new company development trends.

Bogdan Bałkowski / C&T Elmech

5

Abstract

This article discusses the issue of reactive power and higher harmonic compensation by means of clas-sic systems based on capacitor batteries and modern power electronics dynamic systems of the STATCOM and EFA type. The author, through the process of evaluating

advantages and disadvantages, is strongly in favour of the modern dynamic solutions and underlines their benefits and usability in the attempts aimed at improving the quality and efficiency of electric energy usage.

Modern Reactive Power and Higher Harmonic Compensation Through the Utilisation of STATCOM and EFA Dynamic Compensators

MODERN REACTIVE POWER AND HIGHER HARMONIC COMPENSATION THROUGH THE UTILISATION OF STATCOM AND EFA DYNAMIC COMPENSATORS


Modern power systems are often compelled to deal with low energy quality and low energy efficiency pro-blems, in which case higher harmonic and reactive power compensation constitutes one of the most important issues. Even though, technically speaking, this problem is not new, its scale has been growing, which makes it more significant, and the results of low energy quality and low energy efficiency have become more severe. The above is-sues do not only refer to our domestic market. This issue is perceived as a global phenomenon and numerous coun-tries, including the EU Member States, have treated it as very important. However, in Poland, the energy efficiency of the economy is no less than two times lower than the European average. In this perspective, the Act promoting and supporting savings in the final energy utilisation has become indisputably vital1. This Act, to come into force at the beginning of 2011, is the fulfilment of regulations included in the 2006/32/EC European Parliament and Council directive and its main goal is to achieve 9% energy savings by 2016. This means that the problems related to energy quality are accompanied by issues related to energy utilisation efficiency. In such a dynamic economic environment (taking into account the energy prices increase), C&T Elmech suggests taking a comprehensive look at the issue of higher harmonic and reactive power compensation and appropriate new solutions.

In this context it is worth comparing technical capabilities of conventional filter and compensating equip-ment (originating from the first half of the 20th century) and new solutions (originating from the turn of the 20th and 21st centuries), taking into account the advances in the area of power electronics which have been repe-atedly tested in the industry. This paper will discuss only low- and medium-voltage solutions, i.e. ones targeted at industrial recipients. The classic solutions in this range include, in particular, capacitor batteries for reactive power compensation and passive filters for selective higher harmonic compensation.

Capacitor batteries may be divided into groups taking into account the method of their connection to power systems:

• by means of mechanical connectors, i.e. contactors;• by means of static connectors, i.e. thyristors.The first capacitor battery with mechanical connectors was used in 1914 and the first static solutions did

not appear before 1971. In both the above cases, the connectors are used for ON/OFF capacitor batteries con-nection (called compensation level). Taking into account the above, such compensation has a stepped characte-ristic and its precision depends on the value of the compensation level. This means that most usability features of such solutions depend on the type of used connector and compensation level value.

The systems featuring mechanical connectors may be used in power systems with slowly variable loads and the activation of individual levels is usually accompanied by significant current surges and momentary unbalances. Such phenomena are also most often registered in switchgear protection systems. Due to the above, and taking into account the possibility of achieving higher dynamics of the capacitor batteries switching, it is much more beneficial to utilise systems featuring thyristor connectors and modern processor controllers. These systems causing no disadvantages using mechanical connectors are also most often used for follow-up compensation of reactive power. However, they are also limited by several factors resulting, among others, from relatively low semiconductor breakdown voltage.

1 See http://bip.mg.gov.pl/node/11629 for the bill and the justification.

6

Taking the above into account, they are connected directly to the power network with a voltage usually not exceeding 690 V or to power networks with higher voltages by means of a matching transformer. Solutions connected directly to power systems with a voltage equal to 17 kV are also available, however, their prices con-stitute significant economic barriers. Capacitor batteries are sensitive to harmonic voltage present in a power network so they are secured with chokes protecting them against undesirable frequencies. This feature may also be used for aligning with a selected harmonic frequency in order to compensate it.

Passive LC filters are uncontrolled systems activated by an operator. Their design is similar to that of a capacitor battery with mechanical connectors. Moreover, they are usually single-stage filters permanently inclu-ded into the power system. Additionally, the selection of their L and C elements results from the required power and the frequency of compensated harmonics. LC filters also have the reactive power compensation capability because in the case of the basic harmonic (50 Hz), the network “sees” the filter as capacity. When simultaneous filtration of several harmonics is required, several LC passive filters are installed and each of them is tuned to a different required resonance frequency (filtration).

The above description shows that capacitor batteries and passive filters are based on the same L and C elements. This results in the fact that they share numerous features, both advantageous and disadvantageous.

Having analysed the characteristics shown in fig. 2 and properties of passive compensating and filtering systems based on LC elements, the following conclusions may be drawn:

• The compensating system power depends on the square value of the power network voltage fluctu-ations2. Thus, for example, a change of voltage by 10% causes a 21% change in the reactive power in a non-regulated compensating battery. This may also cause load overcompensation and, as a result,

Fig. 1. Approximate characteristic of harmonic LC 5 passive filter dampening

Fig. 2. Voltage-current characteristic of capacitor batteries and passive filters

2 The value of power generated by the capacitor battery is expressed by the following equation Q = U2 wC.


7

further voltage increase in the point where the compensation system is connected to the power ne-twork. When a non-regulated compensation battery or compensator which is not dynamic enough is used, this mechanism may cause instability of the power network including uncontrolled resonance.

• The increase in the power network voltage causes an increase in the compensation capacitor battery current, which in extreme cases, especially in the presence of the network voltage distortions, may result in their overload.

• The passive filter characteristic presented in fig. 1 shows that for frequencies higher than the resonance (damped) frequency an LC passive filter retains the damping properties, however, in the case of lower frequencies this filter is “seen” by the network as capacity, which may result in increased network in-stability in this frequency range.

• Higher harmonic passive filter systems are usually designed for selected harmonic frequencies charac-teristic for a given reception. When analysing fig. 3, one can see that the harmonic spectrum may be variable while maintaining an almost constant THD value. This causes additional problems for correct selection of compensation system passive elements. This is also one of the reasons for understating the estimates for the required power and the number of compensated harmonics and one of the re-asons why the filtration is ineffective and passive systems are overheated. For similar reasons, using simple passive filters, when it is possible that significant inter-harmonic components may be present, is not recommended either.

• Permanent inclusion of passive filters with load variability causes overcompensation (not appreciated in the power engineering industry), which, as mentioned above, creates conditions facilitating reso-nance.

Moreover, due to technical and safety reasons, in classic compensation systems it is rarely possible to achieve a power factor higher than tgø = 0.3. This is approximately 10% of the energy we pay for; however, it is not practically used because it is reactive energy. Nowadays, applicable regulations do not impose such a requirement. It is, however, worth analysing the new energy efficiency regulations which provide favourable conditions for taking actions aimed at a further increase in the tgø coefficient value.

Fig. 3. The figure shows an example variability of the WH voltage spectrum dependent on the control angle of a 6-pulse thyristor converter


8

Does criticism expressed towards the classic compensation systems mean their elimination? Of course not. In the author’s opinion, classic compensating-filtering systems and active systems should

be merged in hybrid solutions utilising the most favourable features of both systems, i.e. by eliminating their disadvantages and focusing on advantages. Hybrid solutions, ensuring much better energy quality and its more efficient usage, are also characterised by an optimum relation between the price of a solution and results achie-ved.

Fig. 4 shows a concept of combining, in one compensation system, the advantages resulting from low pri-ces of LC and SVC compensators, compensating their disadvantages and providing them with positive features of STATCOM (STAtic COMpensator) dynamic compensators and EFA (power active filter), particularly the Xinus Q and Xinus D systems respectively which are manufactured by C&T Elmech.

The discussed concept uses an LC compensator permanently connected to the power network, which, in no load conditions, is compensated by STATCOM facilitating inductive and capacitive reactive power generation and load symmetrisation. The follow-up SVC capacitor battery is a quantified source of capacitive reactive power which ensures rough regulation of reactive power; however, the STATCOM system ensures smooth regulation and high dynamics of the compensations system within the range of one SVC compensator regulation stage.

Fig. 4. Compensation system concept using classic compensation systems with dynamic systems (type Xinus Q and Xinus D)

Cooperation of the three above compensation system elements allows for configuration of a fully regula-ted reactive power source with optimum costs, high dynamics, high speed of operation, smooth operation and resistance to transient conditions present in the power network. It also ensures precise filtration of the current reception harmonics.

Fig. 5a shows an output characteristic of a typical STATCOM system. It is evident that the generated compensating current is totally resistant to voltage fluctuations and the generated reactive power is fully con-trolled. This observation will facilitate the analysis of the simplified characteristic of a hybrid system (fig. 5b). As is evident, the suggested configuration has features characteristic both for classic and active systems. A hybrid compensator still is capable of providing automatic correction of the reactive power generation and facilitates generation of this power with a maximum value two times higher than the STATCOM system alone.

RECE

PTIO

N

LC c

ompe

nsat

or

C ba

tter

y

C ba

tter

y

STAT

COM

STAT

COM

Q power step regulator

Q power follow-up regulator

Q power precise regulator

MASTER power regulator


9

Fig. 5. a) Comparison of IU characteristics of single STATCOM and b) in configuration with classic SVC type systems and LC compensator

Another element of the compensation system is a Xinus D power active filter of a relatively low power value. It facilitates precise and dynamic filtration (of the compensation) of current reception higher harmonics, regardless of its spectral composition and parameters of the power supply network. The Xinus D active filter may also support the STATCOM system operation, i.e. additionally compensate the reactive power and symmetrise the load within its power reserve rage.

The main uses of the suggested hybrid solution are:• energy quality improvement including the guarantee of a high power factor for the power system;• effective voltage stabilisation including the flicker phenomenon reduction; • load symmetrisation and active compensation of higher harmonics;• improvement of distribution lines power efficiency; • improvement of the system transmission capacity including improvement of power system stability

(oscillation damping).

Tab. 1. Comparison of basic features of classic and hybrid systems

Feature Classic system Hybrid system

Time of reaction to reactive power fluctuations 20 ÷ 200 ms 0.25 ÷ 0.5 ms

Resistance to voltage fluctuations in the network none full

Reaction to interferences possibility of resonance full resistance

Capability of generating capacitive and inductive power capacitive capacitive and inductive

Harmonic filtration limited, selective full with changing spectrum

Load symmetrisation none full

Flicker reduction poor effective

This paper is not only a purely theoretical discussion regarding existing possibilities. Even though a sys-tem with such a configuration has not been operated in Poland yet, it is worth analysing the reactive power and higher harmonics compensation system being constructed by C&T Elmech on the main winding machine in the KWK Ziemowit mine. C&T Elmech is the only Polish company which has ensured compensation of the extremely turbulent reception in the drive system of a winding machine. The experience gained will be used for construc-ting a compensation system in another machine. This time, C&T Elmech engineers intend to ensure cooperation of a classic SVC type follow-up compensation system with advantageous features of a Xinus D active power filter. Utilising high dynamics of an active filter and reactive power compensation capability in order to additio-

Vt Vmax

Vt Vmax

BC

ICmax 0 ILmax ICmax ID-STATCOMmax 0 ID-STATCOMmax

Lmax at Vmax

a) b)


10

nally compensate the system is an example of a partial realisation of the concept under discussion. Fig. 6 shows a schematic diagram of the implemented project3 and fig. 7 shows a visualization of the newly designed active filter (power = 2 MVAd).

3 Elektroinfo: nr 12/2007 – Xinus active filters; no 9/2010 (87) – Reactive power and current higher harmonic compensation in medium-voltage networks using hybrid systems based on new XInus active filter solutions.

Fig. 6. Drive system for the northern and southern winding machines with the compensation system. Total reception power Smax = 19 MVA, S.r = 11 MVA, follow-up capacitor battery of power equal to Qbat = 6 MVAr (1 MVAr degree), Xinus D active power filter of power equal to D = 2 MVAd

Fig. 7. Visualisation of an active filter (power = 2 MVAd) with a 6 kV/l,l kV matching transformer


11

It will be the first project of this type implemented in Poland and we are looking forward to its results. Fig.8 shows half of an operational reactive power and higher harmonic compensation system manufactured by C&T Elmech used on the S 1.2 winding machine in the LW Bogdanka coal mine.

Fig. 8. View of an operational Ultra system of an active Xinus D-lMVAd filter and capacitor battery (power = 1 MVar)


12


Damian GadzialskiGdańsk / Poland

He obtained M.A. degree in the Academy of International Economic and Political Relations in Gdynia and engineer’s degree at the Gdańsk University of Technology. He specialises in business development and creating product strategies. He gained his experience in the business field in the Financial Department of Prokom Software SA. He has been involved in power engineering since 2008, firstly, as an assistant to the Chairman of ENERGA SA and next, as the manager for projects concerning implementation of power microgenera-tion ideas on the Polish market. Presently, he holds the position of Product Manager for the following products: Energetyczny Dom (Power House) and Energetyczne Osiedle (Power District) in ENERGA--OBRÓT SA. He is also responsible for evaluation of innovations in the micro-power generation field.

Damian Gadzialski / ENERGA-OBRÓT SA

13

POTENTIAL OF THE PROSUMER MARKET DEVELOPMENT IN THE CONTEXT OF POLISH SYSTEM CONDITIONS


1. PROSUMER MARKET

Bill Gates used to say “Business @ the Speed of Thought”1 and until not long ago one might think that this saying has no relevance on the power distribution market. Meanwhile, the 21st century power system is defined as an intelligent network of harmonised sources, receptions, services and users (prosumers) [1]. This form of power engineering makes it necessary to implement swift qualitative changes in the management systems, technology and legal regulations.

In the new system, a special role must be granted to users whose requirements and needs define its real shape. A prosumer is a recipient who actively participates in the network, additionally becoming a system se-rvices provider [3]. In practice, it means power generation, its sales and influence on the power consumption curve through effective management of receptions, including the creation of own energy tariffs. In the scope of power generation, the prosumer’s activity consists in fulfilling their own energy needs in order to achieve full self-sufficiency.

The basic form of a prosumer on the Polish power engineering market is an individual customer living in an intelligent power house. Electric energy in such houses is generated by, e.g. wind micro power plants and solar cells. In case there is power shortage in the building, the system automatically draws the necessary po-wer from the power network. Larger power recipients adapt their operation in a manner making it possible to utilise maximally large amounts of network power in low-price periods. In cases where surplus energy is gene-rated, the system automatically transmits it to the network. Power consumed and power returned is measured with a two-way measuring meter which communicates remotely with a power company making it possible to balance real settlements. An electric car charger is installed in the garage and its batteries are charged during the low-cost network tariff period and, in an emergency, it may be used as a discharge unit. Thermal power is generated in solar collectors and its surplus is stored in the ground from where it is later drawn by a heat pump. A recuperator retrieves the heat from the ventilation air discharged from the building. On the other hand, the ground heat exchanger uses the ground heat to warm up the air delivered to the building. In order to achieve full energy efficiency, the prosumer’s building is thermally insulated. The technological scope may be much wider than presented in the above example, however, it is not relevant to the essence of this issue, i.e. understanding the characteristics and roles of a prosumer in the new system.

Potential of the Prosumer Market Development in the Context of Polish System Conditions

Abstract

The aim of this paper is to justify the implemen-tation of distributed micro-power generation systems in Poland and present existing system barriers in this field. The paper defines the term and role of a prosumer on the Polish power distribution market and their potential influence on the National Power System. A prosumer is understood as an active power distribution market participant, service provider and co-creator of offers. Their activity chiefly consists in generating energy for own purposes and, if surpluses are achieved, transmitting it to the power grid. The paper describes major obstacles for

the distributed microgeneration development in Poland. For the purposes of this publication, the obstacles have been divided into 3 individual topic groups: power and settlement of account, construction and financial. The barriers presented result from lack of proper adaptation of the national legislation and existing practices in the power sector. Removal of barriers for microgeneration development will require extensive qualitative changes in the Polish power industry which should be beneficial for all involved entities, i.e. customers (prosumers), power distribution companies and the state.

1 Bill Gates, Biznes szybki j@k myśl, Prószyński i S-ka 2001.

14

However, one question remains: what is the prosumer’s influence on the network? It is true that the Polish National Power System is characterised with extensive power output centralisation and its asymmetric distri-bution [2]. So the energy security in Poland depends on the condition of the transmission grid, i.e. capability to really deliver the generated electric power to the end user. In some Polish regions, the transmission is encum-bered with a very high loss coefficient.

One of the solutions to the problems resulting from the above-mentioned conditions is the effect of the prosumer’s presence in the system, i.e. creation of a centralised generation system mostly based on renewable resources. The main argument for introducing the distributed generation is the fact that it complements the central generation on the basis of the complementarity rule. Microgeneration advantages are utilised in an in-creasingly growing extent in such countries as Germany, the United Kingdom, France and USA. This results from the fact that governments of these countries have implemented policies actively supporting such solutions [4]. In Poland, despite very slight government support, there are conditions justifying the usage of microgeneration – see fig. 1 for large power outputs distribution in Poland. The figure shows the disproportion of distribution resulting from the deficit of installed power in the north of the country.

New central outputs are provided as a result of a very long investment process. Additionally, transmission of power from large power plants and delivering it to end users requires constant development and maintenance of transmission grids. At the same time, the power demand grows and dangerously reaches the production vo-lume in peak hours. The Polish industrial power engineering requires bottom-up support, mainly in the scope of flattening the daily electric power demand curve and decreasing the thermal power demand during the heating period. Distributed micro-power generation is a good solution which, in a relatively short time, may start filling in the blank spaces on the power output sources map of Poland.


Fig. 1. Electric power installed in large central sources in Poland Source: http: /www.cire.pl/rynegii/

Installed power

over 1 GW

from 0.5 to 1 GW

from 0.2 to 0.5 GW

up to 0.2 GW

15

The main feature of the distributed microgeneration is its complementarity in relation to the power ma-crogeneration. Its benefits become visible only when they play an optimising and complementary role in relationto the central generation advantages [5].

Tab. 1. Comparison of central and distributed generation. Source: Own work.

Microgeneration Macrogeneration

Essence Distributed energy production for users’ own purposes in their direct neighbourhood

Centralised production of energy in large sources which is later transmitted long distances

Primary energy Wind, sun, gas, waste energy, hydrogen Coal, gas, oil, wind

Efficiency of primary-final energyconversion

Depending on the technology. Generally, the efficiency is lower than in macrogeneration.

Depending on the technology.Generally, the efficiency is higher than in microgene-ration.

Transmission losses Potentially, lower losses resulting from the vicini-ty of end users.

A large amount of energy is lost during transmission.

Failure rate and reliability It lowers the transmission load which results in lower grid maintenance. In the case of a grid mal-function, some sources are still operational and supply the autonomous power microsystems.

It increases the transmission load which results in higher grid maintenance. In the case of grid or source malfunction, very large areas may be cut off from power.

Consumers’ choices A customer is able to choose the supplying energy and does not bear the costs resulting from trans-mission losses and thefts.A customer becomes (at least partially) indepen-dent from the increases in prices of conventional energy media.

A customer is only able to choose a provider and tariffs without any influence on technologies used. Additio-nally, they are charged with costs resulting from grid losses and energy thefts. Additionally, they must suffer the increases in prices of conventional energy media.

Economics The systems will be less expensive when they are produced in large numbers.

More economical taking into account the larger scale of energy generation in a single source.

2. BARRIERS FOR THE PROSUMER MARKET DEVELOPMENT

It is thought that the main barriers for the prosumer market development are the power companies’ acti-vities. However, in practice, this is not true. The main barrier is the legislator who, meeting the requirements of the 3x20 package, seems to forget the fact that the potential of renewable power engineering consists mostly in its distribution. This results from the fact that this power engineering is productively unstable, which results in the logical necessity to plan a distributed and decentralised system. However, the presently applicable desi-gning standards and legal regulations are adapted to a centralised system. They have been created on the basis of the assumption that maintaining the grid stability will be ensured by central and inter-synchronised power plants. For the purposes of distributed microgeneration development, it is urgently required to create a system allowing for the cooperation of the public power system with a large number of distributed sources. Taking into account the character of the barriers for microgeneration, they may be divided into the following types: barriers related to power and settlement of accounts, construction and financial barriers.

As far as the barriers related to power and settlement of accounts are concerned, the requirement of having a concession for electric energy generation created when a prosumer intends to benefit from the Elec-tric Energy Certificate of Origin support system or sell the surplus energy to the network causes the biggestdamage. The legislator issues green certificates only to entities having a concession and gaining this concessionin relation to a microgenerating system (power of a few kilowatts) is only slightly economically justifiable. Ad-ding to the above the necessity to run an economic activity, which results in losing the G2tariff privileges, this solution has absolutely no sense from an economical point of view. The concession requirement relates also to any transactions between a prosumer and a power distribution company. A technically simple system based on

2 Tariff G is a collection of fee rates for recipients consuming power for household and collective residential unit purposes.


16

an intelligent two-way meter becomes unattainable taking into account the stringent regulator’s requirements with regard to a microgenerator. Another barrier is the procedure of issuing the connection requirements by di-stribution system operators. Microgenerators connected to the low-voltage grids are treated equally with large wind farms, which is best confirmed by OSD application forms dividing generators into large and medium ones.The often quoted issue of discharge operation protection was solved a long time ago by modern inverters which have been successfully used in numerous countries. Intelligent grid elements more advanced than the remote read-out facilities are still in the research stage and their implementation requires effective cooperation among experts from numerous power engineering fields.

On the other hand, the construction barriers focus on the requirements of obtaining the construction permit and an element more important than the procedure itself is its cost. If a 1 kW wind power plant is con-structed, the necessity to obtain the construction permit increases the whole investment costs by approxima-tely 30-50 per cent. If we add the risk of the application being rejected and the time necessary to prepare the application, it appears again that the undertaking profitability decreases sharply. The lack of construction lawmatter-of-factness in the microgeneration area results also from the fact that, in its understanding, a wind po-wer plant is both a civil object and a structure, which makes us think that the regulator had big problems making up their mind. While talking about details, it is worth mentioning the fact that there are two ways of installing a wind power plant without a construction permit. Firstly, it is an installation on a mast with stays which is not permanently based on the ground. However, this solution takes a lot of space, is not visually aesthetic and is less safe to some extent. Secondly, the solution may be installed on a building roof. This is possible due to the fact that devices installed on roofs do not require a permit unless they extend over 3 meters above the highest point of the roof. However, such turbine installation practically rules out the possibility of utilising power plants with a horizontal rotation axis, limits the power range to approximately 1 kW and decreases the generation efficiency. The construction barriers also have a significant influence on the heat pumps (particularly when avertical connector is the lower source); however, the regulator’s requirements are slightly more justified in thiscase. In order to install a heat pump bound with the ground in any way, a construction permit is required (it can be included in the construction permit for the whole house). Additionally, according to the permit required by the Water Law Act, an owner of the land may, without a water-rights permit, use the waters located within their premises if the water consumption does not exceed 0.5 m3/h. Generally, a heat pump requires more water which, after passing through its heat exchanger, is treated as wastewater. Activities such as discharge of such amounts of wastewater, using water for energy generation purposes, mixing waters from different aquiferous layers and drilling wells of a depth exceeding 30 m additionally require obtaining a water-rights permit. Another significant limit related mostly to vertical collectors is the necessity to acquire the consent of the National WaterManagement Authority and, in cases where the bore-hole depth exceeds 100 m, also the consent of the State Mining Authority. From the point of view of the Construction Law, only air-water/air type pumps do not require any permits as, in their case, no earthworks are necessary; however, there are interpretations claiming that construction of an external pump on a concrete base may be treated as construction of a device on a foundation permanently bound with the ground.

The last group of barriers is related to obtaining external funding for microgenerating installations. The state and European Union policy explicitly supports enterprises in the field of renewable energy sources con-struction. However, an individual investor intending to implement such an undertaking in the micro-scale enco-unters an impenetrable barrier: namely, bureaucracy. One may agree that preparing applications for large inve-stment projects makes sense, but in the case of a few kilowatt power installations the time and work required for their preparation seems to be disproportionate to the results gained. Another issue is the requirement to provide the application with such documents as a construction permit or environmental impact opinion which, in some microgeneration cases, are simply not required by law. In practice, this may mean the rejection of an application at the formal evaluation stage.


17

REFERENCES

3. SUMMARY

Implementation of microgeneration in Poland is significantly justified from the technical and economic point of view and the market shows a large demand for prosumer solutions. The justification for such solutions becomes even more evident when judging the increasing success of neighbouring countries in the process of implementing microgeneration solutions complementing the central system. However, Poland is totally unpre-pared for such extensive qualitative changes. This is visible in three separate regulatory areas. The law simply does not allow a prosumer to easily implement microgeneration systems, generate energy, obtain benefits re-sulting from this generation and use the renewable power engineering support sources.

The first most important issue in the process of the distributed power engineering evaluation will be the removal of the barriers described in this paper and preparing power distribution companies for operational readiness in relation to organisational challenges resulting from distributed power engineering. Following the examples of western societies, it is additionally worth considering redefining the national power strategy so that it takes into account, even to a minimum degree, the bottom-up activities of future prosumers in relation to the 3x20 package implementation. The next step is to make recipients aware of their possibilities and roles. As a result of the above-mentioned actions, a distributed power plant mostly based on renewable resources may be created which will ensure more independence for the recipients, a new revenue source for power distribution companies, existing market structure optimisation and achievement of environmental and energy safety goals for the state.

1. Gellings Clark W., The Smart Grid: Enabling Energy Efficiency and Demand Response, CRC Press, Palo Alto 2009.2. Gładyś H., Praca elektrowni w systemie elektroenergetycznym, WNT, Warsaw 1999.3. McLuhan M., Take today-the executive as dropout, Harcourt Brace Jovanovich, San Diego 1972.4. Micropower Europe association report, Mass Market Microgeneration in the Europe Union, Brussels, 2010.5. Parker D., Microgeneration: Low energy strategies for larger buildings, Architectural Press, Burlington 2009.

REFERENCES


18


Ksawery OpalaGdańsk / Poland

Graduated from the Faculty of Electrical and Control

Engineering at Gdańsk University of Technology (2001).

Presently, he is working in the Gdańsk Branch of the

Power Engineering Department as a doctoral student. His

main scientific areas of interest include: voltage

automatic control engineering, area voltage and reactive

power control, EE network operation status analysis and

power distribution calculations.

Jacek JemielityGdańsk / Poland

Graduated from the Electrical Engineering Faculty of

Gdańsk University of Technology, M.A. Engineer degree in

automatic control engineering and electrical metrology.

In 1989– 1993, he worked as a doctoral student at

Gdańsk University of Technology where he dealt in the

metrology of non-electrical quantities. Since 1993, he

has been working as a doctoral student in the Gdańsk

branch of the Power Engineering Department. He deals

with the system automatic control engineering for

voltage regulation, algorithm development and their

technical implementation.

Jacek Jemielity / Power Engineering Department, Gdańsk Branch Ksawery Opala / Power Engineering Department, Research Institute, Gdańsk Branch

19Method of Voltage Situation Assessment in the Transmission Grid Taking Into Account

the Regulation Technology Operation

METHOD OF VOLTAGE SITUATION ASSESSMENT IN THE TRANSMISSION GRID TAKING INTO ACCOUNT THE REGULATION TECHNOLOGY OPERATION


1. INTRODUCTION. OVERVIEW OF PRACTICAL METHODS OF VOLTAGE SITUATION ASSESSMENT IN A TRANSMISSION GRID

The majority of system malfunctions, regardless of their causes, develop even further due to the loss of the voltage stability reserve. Taking into account the above, in the 1990s, different voltage situation assessment methods were used and developed with a view to informing the operator about the system stability reserve.

There are several methods and indices for voltage stability testing. They differ as far as calculation me-thods and results obtained are concerned. Selection of a given method may provide results related to voltage stability of individual nodes or area stability. Some methods make it possible to determine the stability reserve and others define the stability reserve indirectly. Numerous methods involve multi-variant calculations and re-quire painstaking analyses of results obtained.

The practically used power network voltage safety assessment methods are: modal analysis, QV and PV nose curves method and continuation method. These methods are based on standard power distribution cal-culations, which means they require an up-to-date model of the transmission and distribution grid. They are time-consuming and only the continuation method being a slight simplification of the QV and PV curves method may be used for determining the system voltage stability reserve in a quasi-online mode. See publication [1] for details.

So far, operators have been practically using several indices facilitating indirect system stability reserve assessment. One of them is the reactive power reserve index (for generators, capacitor batteries, etc.) used, among others, in the BPA system in North America [2]. This index ensures ongoing monitoring of the summary range of reserves available in the system. A similar indicator called the Voltage Stability Index (VSI) has been used in the Italian ENEL system [3]. This index has been complemented with the factor taking into account the derivative of changes in the available reactive power reserves.

(1)

where:qi(t) – momentary level of reactive power generation ρ – weight coefficient

Abstract

Selected functions of the Area Voltage Regulation System (SORN) have been implemented in the Bydgoszcz Power Dispatch Centre (ODM) where the method of volt-age situation assessment described in this paper has been utilised. The method of automated control over the ARNE and ARST regulation systems operation elaborated in the SORN system aims at supporting dispatchers during the ongoing regulation status assessment, danger conditions detection and counteracting the system malfunction re-sults. This aim has been achieved by utilising the method of voltage situation assessment in the transmission grid which used normalized numerical indices describing the regulation nodes operation point. Next, the analysis of

trends related to changes in these indices was used for detection with time advance of adverse changes in the voltage profile or reactive power reserves level. The described method employs the fuzzy inference system in order to generate prompts for Power Dispatch Centre (ODM) dispatchers. The limited SORN system functions fulfilled presently are to enhance the power system safety using the existing ARNE and ARST regulation systems in-frastructure. Together with the latest modifications of the ARST system algorithms, the SORN system’s basic duty is to prevent worsening or accelerating a voltage malfunc-tion, which results from transformer regulation not being blocked on time.

ttq

tqtVSI iii

)()()(

20

Such indices are individually assigned to each Italian grid subarea for which the so-called pilot node is determined, i.e. a reference point to the area voltage regulation. Thanks to this, the local reactive power mana-gement characteristics have been taken into account.

Another interesting indicator for the approximate stability reserve assessment is the voltage index Li [4] calculated for each grid node:

(2)

where:Vi – voltage of node i calculated for the up-to-date grid modelV0i – voltage of node i with power distribution for zero intakesIt does not inform about the value of the voltage stability reserve; however, it shows individual nodes suscep-

tible to losing stability. In order to calculate the index L, a grid model is necessary but the calculations are simplified,i.e. there is no problem related to the additional load distribution as is the case in the nose curves method.

2. ASSUMPTIONS FOR THE VOLTAGE ASSESSMENT METHOD IN THE TRANSMISSION GRID INTEGRATED WITH REGULATION TECHNOLOGY

The method described in this paper has been implemented in the SORN system taking into account the specific character of the northern National Power System (KSE) area subordinated to the Bydgoszcz Power Di-spatch Centre (ODM). The transmission grid in this area has nominal voltage equal to 400 kV and 220 kV. There are two pumped storage power plants (Żarnowiec and Żydowo) operating in the grid, the HVDC circuit and 11 high-voltage power stations. Both power plants are equipped with ARNE regulation systems. 10 high-voltage stations are equipped with ARST systems and 4 stations are equipped with manually operated capacitor bat-teries. From the point of view of the operator’s practice, the characteristic feature of the area is the intensive utilisation of regulation systems resulting in frequent daily corrections of setpoints set for regulators.

Lack of ongoing availability of the full grid model for the Bydgoszcz Power Dispatch Centre (ODM) (lack of measurements from the 110 kV grid) limits the practical usage of known methods of calculating the voltage stability reserve in the direct online mode. Bearing in mind the fact that only one Żarnowiec power plant may practically influence the ODM area voltage level, the type index (1), taking into account the reactive power re-gulation reserve for this power plant, is a natural method of the ODM area voltage situation assessment. While designing the SORN system for the Bydgoszcz Power Dispatch Centre (ODM), the main emphasis was put on the automatic supervision of the voltage regulation systems operation in order to enhance their operating safety in the case of a risk of a voltage type malfunction.

3. VOLTAGE SAFETY INDICES

In order to facilitate an objective assessment of values monitored within the SORN system, three non--dimensional voltage safety indices have been suggested:

1. WQ – index of reactive power regulation range utilisation in power plants2. WU – voltage index for generating and transmitting nodes defining the deviation from the desirable

voltage profile3. WZ – index of the transformer voltage transformation ratio regulation range utilisation.The operating secondary regulation systems, according to assumed regulation criteria and settings, auto-

matically react to load variations and disturbances in the power system operation. Indices WQ , WZ and WU should facilitate the assessment both of the ongoing value of monitored quantities and the predicted changes in these values on a time horizon from a few to over a dozen minutes. Thus, the currently registered index changes and the resulting change trend are important. Safety indices, which are calculated taking into account the change trend reach, in advance, the W = 1 or W = -1 criterion values thus allowing time for a proper reaction of the dispatcher.

Treating the index run, determined on the basis of momentary values of monitored quantities, as a random process, the index W value, taking into account the change trend, is calculated as a sum of two components:

i

ii V

VL

0


21

W = w(t) + Δw(t) (3) where:w(t) – estimated index value on the basis of the run from moment tΔw(t) – value increase resulting from the continuation of the change trend to moment t, as long as this

tend exists, i.e. has a sufficiently high determination coefficient.Since component w(t) in formula (3) is an estimated value, it causes averaging and smoothing of the index

W run and makes it resistant to momentary disturbances. On the other hand, the incremental component Δw(t) represents the dynamics of changes, including that caused by the regulation system operation.

The experiences resulting from the SORN system operation show that the trend must be determined with a variable time horizon adapted to the shape of the run. We assume that in (3) the predicted increment Δw(t) will be present for the time interval for which the most reliable trend has been determined. Fig. 1 shows two runs: voltage test run (Voltage U) and predicted value of this voltage determined on the basis of analysis of the last 20 minutes of the test run (U + trend). The runs shown in the figure illustrate the operation of the trendanalysis algorithm used.

In sections A-B and C -D in the figure, this algorithm uses an increasing time interval as a basis for cal-culating the change trend. Thus, the voltage change increases. It guarantees detection of slow signal changes lasting a dozen or so or several dozen minutes. In the B-C and D-E sections, following the incremental change of the test run derivative, a quick correction is noticed, i.e. the algorithm finds a more reliable trend determinedon the basis of only the last test signal samples.

3.1. Index of reactive power regulation range WQ utilisationThe phenomenon of voltage instability is often, although not always, accompanied by the earlier depletion

of the generating units reactive power regulation range. Reactive power reserves level monitoring is a basic function of voltage stability monitoring systems. The suggested index of reactive power regulation range utilisa-tion is a standard value of reactive power generation relating to the minimum reactive power Qmin or maximum reactive power Qmax of generating units operating on connected systems of switchgear busbars – 400, 220 or 110 kV. The regulation range centre is always suggested as a reference level. The value of index WQ is definedby the formula (4):

WQ = q(t) + Δq(t) (4)

Fig. 1 llustration of the tend analysis algorithm operation

Method of Voltage Situation Assessment in the Transmission Grid Taking Into Account the Regulation Technology Operation

Voltage U U + trend

Time [s]

Volta

ge [

kW]

22

where:q(t) – estimated ongoing value of the standard reactive power generation, i.e. value Q(t) in the generating

node, related to the difference between the power limit value Qmin. or Qmax. and half of the regulation rangeΔq(t) – predicted index increase provided the existing trend is continued:

(5)

Tab. 1. Criterion values of index W

Value WQ -1.0 0.0 1.0

Node status Maximum utilisation of the regulated capacitive power range Q = Qmin.

Centre of the reactive power regulation range(Qmin. , Qmax)

Maximum utilisation of the regulated inductive power range Q = Q

3.2. Voltage index WUFor the transmission grid, in given conditions an optimum voltage profile is assumed – the voltage set for

the secondary regulation system in connected busbar systems. The suggested voltage index WU is a standard node voltage deviation from the current set voltage related to the voltage of the overvoltage or undervoltage interlock of the regulation system. The advantage of this solution consists in connecting the calculated value of WU with current settings (set voltage) and parameters of the regulation system (interlock).

WU = u(t) + Δu(t) (6)

where:u(t) – estimated ongoing value of the standard voltage deviation related to the value of the closest voltage

interlockΔu(t) – predicted index increase provided the existing trend is continued:

(7)

(8)

(9)

U – voltage of connected busbar systems Uset – set voltageU BLP – undervoltage interlock voltage UBLN – overvoltage interlock voltage

Tab. 2. Criterion values of index WU

Value WU -1.0 0.0 1.0

Node statusVoltage deviation on the undervoltage interlock level

Lack of voltage deviation from the regu-lation system set value

Voltage deviation on the overvoltage interlock level

)()(for

)( )()()( .minmax2

1

.minmax21

minmax21

QQtQQQQQtQtq

.

set

setBLN

set )(for)()( UtUUUUtUtu

set

BLPset

set )(for)()( UtUUUtUUtu

set)(for0)( UtUtu


23

For the transmission grid, admissible regulated voltage limits for the switchgear are assumed. The ma-jority of ARST transformer regulation systems control the voltage in the 110 kV grid. The voltage index for the 220 kV or 400 kV side of a transformer must be determined by calculating the set value of the 110 kV side pro-portionally in relation to the admissible set voltage values (fig. 2).

If the regulation system controls the transformer reactive power flow or the automatic regulation is bloc-ked and done manually, in order to calculate index WU, for the 110 kV side the voltage equal to the measured voltage is assumed as if the voltage regulation has just finished. In this case index WU = 0 (unless the possible change trend is taken into account).

3.3. Transformer regulation range utilisation index WZThe suggested transformer regulation range utilisation index WZ is an auxiliary index which consists in

facilitating the objective assessment of the transformer automatic regulation influence on the current voltagestatus. Index WZ is a normalized deviation of the transformer tap switch location from the “middle” tap by which the nominal voltage transformation ratio is defined. Its criterion values WZ = 1 and WZ = -1 are reached in the extreme transformer switch taps for the minimum and maximum voltage transformation ratio, regardless of the transformer construction1.

WZ = z(t) + Δz(t) (10)

where:z(t) – estimated value of the standard deviation of the tap from the middle location related to the extreme

tap positionΔz(t) – predicted index increase provided the existing trend is continued:

Fig. 2. Method of calculating the 110, 220 and 400 kV set voltage to determine index Wu

1 Two kinds of tap switches are used: the voltage ratio increases or decreases together with the increasing tap number.

Uset. min.Uset. max.


Uset.

Uset.

Uset.

Uset.

24

(11)

(12)

z(t) = 0 for Z(t) = Z0 (13)

where:Z0 – middle tapZmin., Zmax – extreme taps

Tab. 3. Criterion values of index Wr

Value WZ -1.0 0.0 1.0

Tap status Extreme tap of the minimum trans-former 110kV side voltage

Middle tap number, i.e. neutral posi-tion of the transformer tap switch for the nominal voltage transformation ratio

Extreme tap of the maximum transformer 110kV side voltage

Index WZ must be determined excluding shorted taps for which the transformer voltage transformation ratio does not change.

3. 4. Node status monitoring by means of indicesUsing standard indices related to regulation systems facilitates the node status analysis. It is noticeable

that:If the generating units regulated power Q reserve is present, i.e. -1 < WQ < 1, voltage index WU for

a switchgear will be equal to zero or will be close to zero (voltage deviation within the regulation dead band limits). If power Q reaches the limit value Q min. or Qmax, voltage index WU of the switchgear with generation may start changing.

If the regulation system receives a new setpoint, a given index WU for the switchgear will be temporarily different from zero. If, at the same time, other indices, i.e. WQ WU or Wz do not reach criterion values, it means that the regulation system is during regulation (activation delay).

In the case of a station with a 220/110 or 400/110 kV/kV transformer, one may notice that:• if the regulation system regulates the 110 kV voltage, index WU/110 for the 110 kV switchgear will be

equal to zero (deviation within the dead band zone limits) until the extreme tap is reached if regulation is blocked due to, e.g. too low 220 kV or 400 kV voltage. In such a case WZ reaches the criterion value 1 or WU/220 or WU/400 value -1;

• assuming the criterion values by indices WU/220 or WU/400 and Wz for a transformer station may constitute the signal to block the ARST system automatic regulation, change a criterion or correct the setpoint.

The above conclusions enabled the authors to create the Fuzzy Inference System (FIS) generating prompts for Power Dispatch Centre dispatchers. A system built on the basis of fuzzy logic and fuzzy implications tables is often used to create expert systems and knowledge databases. See below for necessary theoretical bases of the method and its practical application in the SORN system.

0

0

0 )(for)()( ZtZZZZtZtz

max

0

min.0

0 )(for)()( ZtZZZtZZtz


25

4. FUZZY INFERENCE SYSTEM BASED ON INDICES WU, WQ AND WZ

The bases of inference used by the authors in relation to the Bydgoszcz Power Dispatch Centre (ODM) area are safety indices WU and WZ values calculated on the basis of data from ARST systems and the value of 400 kV voltage measurement in the Słupsk power station. Input signals (prompts) relate to the set voltage values U110 set (for transformers operating in the 110 kV voltage regulation criterion) or 110 kV side voltage (for transformers in manual regulation) or activation/deactivation of a battery section.

For power plants in the Power Dispatch Centre area and in its direct vicinity, WQ indices monitoring has been used. In the Żarnowiec power plant the value of the index has been used to generate prompts related to utilisation of the HVDC circuit capacitor battery in the Słupsk power system.

4.1. Theoretical bases for the fuzzy sets method in control and expert systemsFuzzy systems are automatons using fuzzy logic laws in order to take decisions in uncertain conditions,

e.g. in the case of lack of a precise mathematical model for the examined phenomenon or if the model is too complex [5]. Such automatons have a certain knowledge database in the form of a set of inference rules which come from an expert who creates the system. The effectiveness of the fuzzy inference system depends chiefly on the quality of the expert’s knowledge and secondly on the correctness of its modelling by means of fuzzy logic.

A typical fuzzy inference process takes place in three stages:1. Making the input values fuzzy, i.e. conversion of real values of input signals into linguistic variables.2. Using fuzzy implications, i.e. analysis of the pre-defined set of relations between fuzzy linguistic terms of

input signals and the output signal.3. Specifying (defuzzification) the output value, i.e. conversion of the blurred value of an output signal into

a given value. Fig. 3 shows a substantially simplified schematic diagram of the fuzzy inference system used in the SORN

system. We have here three input linguistic variables WQ, WU and WZ which are processed into fuzzy sets consti-tuting premises for inference rules included in the database of rules. In the output, we obtain output linguistic variables, i.e. system prompts:

• analogue (suggested 110 kV voltage);• two-stage (warning and alarm signalling and discrete prompts for capacitor batteries control and swit-

ching ARST systems operation modes).

The fuzzy inference system generates a single number in its output, i.e. the result of the input signals specifying block operation. The authors have attempted to obtain the effect of reconstructing the dispatcher’s intuitive actions. The system operation has been tested on real runs registered during one week before the well known malfunction on 26.06.2006 and during the day of its occurrence.

Fig. 3. Schematic diagram of the fuzzy inference system

If..., then...If..., then...If..., then...

...

...

...If..., then...


Making inputs fuzzy

Inference rules database

Specifying outputs

Warning Signal

Capacitor Banks Operation

26

4.2. Making input signals fuzzyIn the case of the Wu index values of the lower and upper transformer voltage (W110 and W220/400 respecti-

vely) five linguistic expressions have been assumed which define the index value ranges: very low, low, medium, high and very high. Fig. 4 below presents the described expressions and their respective WU index ranges as functions of fuzzy sets affinity corresponding to the linguistic variable describing index WU.

In the case of safety index WQ of the power plant, three linguistic expressions describing the index value ranges have been assumed (fig. 5): low, medium and high.

One must pay attention to the fact that, thanks to assuming a proper form of the safety indices related with the interlocks and regulation systems setpoints, the inference system input signals become permanently fuzzy for changeable system operation conditions. For example, assuming a new voltage profile for an incomple-te grid configuration naturally causes the WU indices re-scaling and the prepared inference rules remain valid.

Fig. 4. Example of the index WU affinity function Fig. 5. Example of the power plant WQ index value becoming fuzzy

2 Criterion D – voltage regulation of the transformer lower voltage side 110 kV.

4.3. Inference rules for a transformer stationFor transformers working in the transmission grid, when the ARST regulation system works with criterion D2,

a rule has been assumed stating that the lower set voltage 110 kV – Uset is the control value. If the control value regulation system is deactivated, it is simply the 110 kV – U110 voltage. Following the process of defining the output signal a prompt adequate to the ongoing ARST system operation status is generated, as long as realisation of this prompt requires changing the location of the tap switch. Tab. 4 shows the set of rules depending on the W110 and W220/400 input variables.

Tab. 4. Inference rules used for defining prompts related to UDzad. for a typical transformer station

W220/400W110

very low low medium high very high

very low 0 0 2 2 2

low 0 0 1 1 2

medium -2 -1 0 1 2

high -2 -1 -1 0 0

very high -2 -2 -2 0 0

In the process of defining the output signal, values -2, -1, 0, 1, 2 are converted into the set voltage from the <Uset min., Uset max > range (fig. 2).


very low low medium high very high low medium high

WU

27

4.3.1. Transformer station – recognising emergency conditionsThe simplest emergency condition detection in a power grid may be achieved by means of tab. 5. See

below for W110 and W220/400 coefficient combinations explicitly predicting voltage problems in a given transformerstation and its immediate vicinity.

Tab. 5. Inference rules used for defining prompts related to emergency conditions for a typical transformer station

W220/400

W110


very low-2 0

low

medium -1 0 1

high0 2

very high

Meaning of output signals:

– 2 area of voltages which are too low showing the power deficit in a given supply point

2 area of voltages which are too high showing the power surplus in a given supply point

– 1 area of voltages which are too low showing a possible approaching system malfunction

(voltage collapse)

1 area of voltages which are too high showing a possible approaching system malfunction

0 area of ARST control engineering standard operation in a given station

0 area where ARST condition is ambiguous (e.g. following a setpoint change).

4.3.2. Transformer station with a capacitor bankIn the case of a station with a capacitor bank installed, the set of input values included in tab. 4 is still

valid. In tab. 6 additional inference for capacitor banttery activation or deactivation has been introduced.

Tab. 6. Inference rules for a transformer station with a capacitor battery

W220/400

W110


very low 2 2 2 0 0

low 2 1 1 0 0

medium 2 1 0 -1 -2

high 0 0 -1 -1 -2

very high 0 0 -2 -2 -2

Meaning of output signals:2 – activate all sections1 – activate a section0 – no changes– 1 – deactivate a section– 2 – deactivate all sectionsThe prompt related to the capacitor bank control is valid as long as the input signal values do not change.

In this case, the regulation effect is immediate and should influence the inference rule input signals.


28

4.3.3. Słupsk 400 kV station (SLK – a special case of a station equipped with a capacitor batteryThe SLK station differs from other stations mostly in the fact that the HVDC circuit is connected to its

400 kV switchgear and is not equipped with the ARST regulation system. Indices W110, W220/400 and Wz are not calculated for this station. Taking the above into account, a separate fuzzy inference table has been developed. The capacitor battery operation in SLK (2 x 95 Mvar and 95 Mvar filter) is strongly correlated with the Żarnowiecpower plant operation (reactive power generation). Due to the above, the variable input voltage has been defi-ned as the voltage in the 400 kV switchgear in the SLK station and index WQ for the Żarnowiec power plant.

The capacitor battery activation or deactivation moment must take into account the HVDC circuit opera-tion status. While importing or exporting power, per each 200 MW of power flowing though the HVDC circuit,approximately 80 Mvar of reactive power is reserved. If power is not transmitted through the HVDC circuit, a battery operation may be controlled as in other typical stations. Taking the above into account, following the filter activation, practical voltage regulation with both batteries is possible only by power transmission in the-200 MW < PHVDC < 200 MW range. Prompts are not generated for other transmission values. On the basis of operational experiences, it has been assumed that if the SLK400 > 419 kV level is achieved, one battery is deactivated, or if none of them is operational, a choke is activated. However, when voltage SLK400 < 391 kV, one battery is activated. Additionally, before a battery is activated, the set voltage in the Żarnowiec power plant is usually decreased by a few kV.

See tab. 7 for inference rules assumed for capacitor battery operation control in the SLK station.

Tab. 7. Inference rules for capacitor battery operation in SE SLK

Item. SLK – HVDC SLK400 ZRC – WQ Resuly

1

-200 MW < PHVDC <

200 MW

low

low

activate battery

2 medium no changes

3 high deactivate battery

4 low

medium

activate battery

5 medium no changes

6 high deactivate battery

7 low

high

activate battery

8 medium activate battery

9 high no changes

4.3. 4. Stations adjacent to power plantsTwo stations this type, ZRC and ZYD, operate in the Bydgoszcz Power Dispatch Centre (ODM) area. Both

stations have all generating units connected to one switchgear. In ARNE systems, the reactive power control is achieved by setting the voltage in the switchgear to which the generation is connected. In the ZRC station, the ARNE system maintains the UG - 400 kV voltage and in the ZYD station the voltage is UD - 110 kV. Set voltage for both ARNE systems is not maintained only after the generators reach the reactive power generation limits. Taking the above into account, monitoring of WQ indices of both power plants has been suggested. These indices define the predicted regulation range and provide sufficient information for the power plant operation asses-sment from the point of view of the reactive power generation. In the SORN system, the indices are presented on a display with two pointers (fig. 6). The black pointer shows the current index value and the red one showsthe WQ. value resulting from the trend calculated on a current basis.


29

REFERENCES

5. SUMMARY

The presented method of voltage situation assessment in a transmission grid makes it possible to identify emergency conditions and generate prompts for dispatchers in order to minimize the malfunction results. The unquestionable advantage of this method is the possibility to use it on the basis of measurements from the transmission grid without the need for creating an up-to-date grid model. Another advantage is the fact of using non-dimensional safety indices taking into account, firstly, the status of the ARNE and ARST regulation systems operation and settings and, secondly, the trend of changes taking place in the power system. The method has been used in the SORN system operating in the Bydgoszcz Power Dispatch Centre (ODM). Analysis of the me-thod results, one year after its implementation, has enabled the authors to confirm its effectiveness. Results of precise analyses of registered data relating to the SORN system operation over a longer period of time will be presented later.

1. IEEE/PES Power System Stability Subcommittee Special Publication, Voltage stability assessment, procedures and guides, January 2001.

2. Taylor CW., Ramanathan R., BPA Reactive Power Monitoring and Control following the August 10, 1996 Power Failu-re, Invited paper, VI Symposium of Specialists in Electric Operational and Expansion Planning, Salvador, Brazil, 24-29 May 1998.

3. Corsi S., Pozzi M., Bazzi U., Moncenigo M., Marannino P, A simple Real-time and on-line voltage stability index under test in Italian Secondary Voltage Regulation, Paris, Cigre session 2000.

4. Hatziargyrion N.D., Cutsem T. van (editor), Indices predicting voltage collapse including dynamic phenomena, tech-nical report TF 38-02011, Cigre session 1994.

5. National Instruments, LabView – PID and Fuzzy Logic Toolkit User Manual, June 2009.

Fig. 6. WQ values display in the ZRC station


Q power regulation range utilisation: Value Trend

30

Janusz Tadeusz CieślińskiGdańsk / Poland

Head of the Department of Environmental Engineering and Combustion Engines at the Faculty of Mechanical Engineering, Gdańsk University of Technology, Associate Professor at the State School of Higher Professional Education in Elbląg. Expert in thermodynamics, heat exchange and exchangers, multiphase flows, process equipment, unconventional energy sources, unconventional energy conversion systems and environmental engineering.Since 2002 editor of scientific publications at the Editorial Committee of Gdańsk University of Technology’s Publishing House. Member of Thermodynamics Section, Committee on Thermo-dynamics and Combustion, Polish Academy of Sciences, and Fluid Mechanics Section, Committee on Mechanics, Polish Academy of Sciences. Scienti-fic consultant of “Technika Chłodnicza i Klimatyza-cyjna. Niekonwencjonalne Źródła Energii” journal. In 2002–2008 held the post of Vice-Dean for Science at the Faculty of Mechanical Engineering, Gdańsk University of Technology. Supervisor of six completed doctoral theses. Contact person for the ERASMUS programme carried out with TU München, TU Berlin, Universität Erlangen-Nürnberg and FH Stralsund.Chairman, secretary or member of 22 organizatio-nal committees of national or foreign scientific conferences. Participated in research projects in numerous foreign research establishments. Author or co-author of more than 200 scientific publica-tions.

Krzysztof KrasowskiElbląg / Poland

Graduate of the State School of Higher Professio-nal Education in Elbląg (2003) and the Faculty of Mechanical Engineering, Gdańsk University of Technology (2006). In 2009 awarded doctorate in engineering at the Gdańsk University of Technolo-gy, specializing in heat exchange and heat exchangers. In 2004–2008 employed at the Faculty of Mechanical Engineering, Gdańsk University of Technology. In 2008 hired by Energa Kogeneracja Sp. z o.o. in Elbląg as Chief Operational Engineer. In 2010 appointed Director for Investment and Development.

Krzysztof Krasowski / ENERGA Kogeneracja sp. z o.o., ElblągJanusz Tadeusz Cieśliński / Gdańsk University of Technology


31

INVESTIGATION OF BOILING HEAT TRANSFER IN AN ELECTRICALLY HEATED TUBE BUNDLE

Krzysztof Krasowski / ENERGA Kogeneracja sp. z o.o., Elbląg Janusz T. Cieśliński / Gdańsk University of Technology

1. INTRODUCTION

Boiling in tube bundles can be observed in multiple industrial devices, like vaporizers, boiling pots or wet boilers. Boiling also occurs in evaporators of absorption chillers used in polygeneration systems, which simultaneously generate heat, electricity and chilling power – by operation of a compression refrigerator or an absorption chiller.

Previous research on boiling occurring in tube bundles had indicated that thermal characteristics of indi-vidual tubes within a bundle are significantly different from values observed for a single tube due to the strong effect of steam bubbles generated at lower tube rows. In addition, a significant influence of bundle geometry was observed, including the impact of the exchanger’s jacket and hydraulic properties of the flow on heat trans-fer within the bundle. There is not much information available concerning local heat transfer coefficients for individual tubes of a bundle.

Marto and Anderson in their research [1] investigated a bundle of 15 copper pipes of which only 10 were heated. The ratio between tube spacing and tube diameter was 1.2. The investigated liquid was R113 refrige-rant. Experiments were carried out at atmospheric pressure. The researchers determined that the lowest heat exchange coefficient occurred at the bottom tube row, and that those lowest tubes intensified heat exchange at higher rows only when the heat flux values were low. The intensification coefficient for the bundle depended on the heat flux density and number of tubes within the bundle. Research also showed that the greatest impact on intensification of heat exchange of a pipe had the pipe located directly underneath. This is described by the so-called bundle effect factor [2], defined as the ratio between the heat exchange coefficient for the topmost tube of the bundle to the heat exchange coefficient of the same tube when the tubes located beneath are not heated.

Qiu and Liu [3] investigated the influence of tube spacing and pressure values on intensification of heat transfer in a bundle of 18 (3 × 6) polished copper tubes with a diameter of 18 mm each, heated with cartridge heaters. The tube layout was staggered. The researchers discovered that intensification of heat transfer in the bundle was highest for a spacing of 0.3 mm. For 0.3 mm spacing, location of the tube within the bundle did not affect the recorded heat transfer coefficients. For spacing higher than 1 mm the influence of pressure on transferred heat was negligible.

Gupta et al. [4] investigated the water boiling process at ambient pressure on smooth electrically heated tubes made of stainless steel, with a diameter of 19.05 mm. Tubes – two or three – were located one above the other, with the spacing (s) to diameter (d) ratio ranging from 1.5 to 6.0. Density of mass flow of water at satu-ration point into the tank G varied between 0 and 10 kg m-2 s-1. This allowed simulating both pool boiling (G = 0) and vaporizer (G > 0). The researchers determined that the heat transfer coefficient at the lowest tube did not depend on the presence of the tubes above. Additionally, the highest heat transfer coefficient (100% higher

Abstract

The paper presents the results of experimental research on water, methanol and R141b refrigerant bo-iling in a horizontal bundle of smooth tubes supposed to represent part of a flooded evaporator. Experiments were carried out for a bundle of 19 tubes in triangular layout for two spacing-to-diameter ratio values – 1.7 and 2.0 – in atmospheric and subatmospheric pressure conditions.

Heat transfer coefficient values – both local, tube-specific, and general, for the entire bundle – were determined. The boiling process at the bundle was visualized with a CCD camera and laser sheet technique.

A correlation equation allowing the calculation of the average heat transfer coefficient for boiling in smooth tube bundles has been proposed.

Investigation of Boiling Heat Transfer in an Electrically Heated Tube Bundle

32

than for the lowest tube) was observed at the topmost tube in the three-tube system, and – interestingly – for pool boiling (G = 0).

Kumar et al. [5] investigated distilled water boiling in subatmospheric pressure conditions (35-97.5 kPa) on two smooth copper tubes with a diameter of 32 mm located one above the other, heated with cartridge he-aters. When comparing their results with those obtained by other authors no significant impact of tube materialon heat exchange coefficients can be observed.

Leong and Cornwell [6] tested an electrically heated bundle composed of 241 tubes. The test setup simu-lated a flooded evaporator. Research was carried out with R113 refrigerant at atmospheric pressure for tubeswith an external diameter of 19.05 mm and spacing (in rectangular layout) of 25.4 mm. Heat transfer coefficientsat the bundle bottom turned out to be close to those observed for a single tube, but at higher rows the values were considerably higher.

Gupta [7] carried out experiments of heat exchange for distilled water boiling at atmospheric pressure in a bundle consisting of 15 tubes (3 × 5) in linear layout, with spacing-to-diameter ratio s/d = 1.5 placed in a large volume tank. Experiments were carried out for heat flux densities between 10 and 40 kW/m² and mass flowdensities 0-10 kg/(m2s). The bundle was made of stainless steel tubes with a diameter of 19.05 mm and effective length of 190 mm each; the tubes were heated electrically.

The results obtained by Gupta show clear differences in heat exchange coefficients for individual tuberows – with the lowest value for the bottom pipe layer and the highest for the topmost layer. The maximum value of heat exchange coefficient for the topmost tube of the central column was seven times higher than the heatexchange coefficient for the lowest tube of the same column, at volume boiling and the same heat flux density of23 kW/m². Similar variations in heat transfer coefficient were obtained by Gupta for tubes in side columns, butin that case the increase of heat exchange coefficient between the bottom and top tube did not exceed 300%.

Dual-phase flows which occur during boiling in a tube bundle depend on many factors, including heat fluxdensity, liquid properties, type of tube surface and layout and spacing of tubes. Nevertheless the heat transfer coefficient for a tube bundle is usually higher than that for a single pipe in the same conditions. This pheno-menon is described by the so-called bundle factor [2] defined as the ratio between the average heat transfercoefficient for a bundle and the average heat exchange coefficient for a single tube.

Webb et al. [8] point out that the heat transport mechanism in flooded evaporators is different than invaporizers. This results from a different tube layout used in flooded evaporators, which restrict natural convec-tion, as well as from the fact that the steam aridity degree at the inlet to a flooded evaporator might even reach15%. It is also extremely difficult to apply results of experiments involving specific tube bundle, type of liquid andprocess conditions for a different case. Attempts to create a theoretical description of this process or model it encounter numerous obstacles. Therefore, calculations are usually based on various empirical or semi-empirical correlations, involving constants which need to be experimentally determined.

A key issue for calculating heat transfer coefficient is to correctly determine the outside surface tubetemperature. This measurement, extremely difficult to accomplish, is influenced by the design of the heatingsection, as well as structure of power supply and control system, and data acquisition method. Therefore, this study presents a system structure which allows simultaneously reading and recording measurement results for boiling in a horizontal tube bundle, characterized by high accuracy of both power control and measurements. Based on experimental data, some of which is presented in this study, a correlation equation has been proposed. It expresses the Nusselt number as a function of boiling number, Prandtl number and geometrical parameters of a tube bundle and allows one to calculate the average heat transfer coefficient for a boiling process in a bundleof tubes with porous coatings.

2. LABORATORY SETUP

The laboratory setup for the investigation of a boiling process in a tube bundle consists of six main sys-tems: modelled tube bundle, experimental tank, power supply and control unit, cooling system, data acquisition system, and visualization system. The cylindrical tank supposed to simulate the jacket of a real evaporator has a diameter of 0.3 m, length of 0.3 m and is made of stainless steel. It is equipped with three inspection windows


33

– in the front wall (with a diameter equal to the jacket diameter) and at both sides (200 mm in diameter). They enable direct observation and visualization of the boiling process. The water-steam system is hermetic and al-lows carrying out the investigation in absolute pressure range 3-300 kPa.

Measurements made with this setup allow determining the average value of the heat transfer coefficientfor the entire tube bundle, heat transfer coefficient for any individual tube of the bundle, peripheral temperatu-re distribution for each tube and enable visualization of boiling structures in the tube bundle. Fig. 1 presents a schematic diagram of the setup and Fig. 2 presents its real-life view.

L1 L2 L3 0

3x230V AC

1

2

3

47

8

9

13

12

11

10

16

17

18

19

56

20

21

23

14

Cooling water inlet

15

22

Cooling water outlet

Fig 1. Diagram of the experimental setup. 1 – Tank, 2 – Tube bundle, 3 – Condenser, 4 – Pressure transducer, 5 – Vacuum gauge, 6 – Relief valve, 7 – Pressure control valve, 8 – Working fluid drain valve, 9 – Cooling water isolation valve, 10 – Flow control valve, 11 – Flowmeter, 12– Cooling water temperature indicators, 13 – Temperature indicators in the tank and tube bundle, 14 – Preheater, 15 – Preheater’s tempe-rature indicator, 16 – MPS power measurement unit, 17 – Power controllers, 18 – CCD camera, 19 – Laser sheet, 20 – Adjustable support of the laser sheet system, 21 – Adjustable support of CCD camera (3D), 22 – Auxiliary light sources, 23 – Data acquisition system.


34

Fig. 2. Experimental setup. 1 – Experimental tank, 2 – Condenser, 3 – Vapour system, 4 – Condensate return system, 5 – Rotameter (cooling water flow measurement), 6 – Cooling water inlet into condenser, 7 – Cooling water outlet from condenser, 8 – Working agent drain valve, 9– Multiplexer, 10 – Reg1-Reg20 regulator panel with power supply switches, 11 – RPG1-RPG20 precise power controllers panel with voltage drop measurement switches AU, 12 – V1-V4 voltmeter panel with MPS network parameter measurement unit, 13 – Adjustable support for CCD camera, 14 – CCD camera, 15 – Laser sheet, 16 – Tubular mainframe of the setup, 17 – Tank support beams, 18 – Vacuum meter, 19 – Ice zero, 20 – Data acquisition computer.

The experimental tube bundles consisted of 19 tubes in triangular layout, with a spacing-to-diameter ratio of 1.7 and 2.0 (Fig. 3a). At one end the tubes were fixed in a punched plate made of stainless steel. The other endswere free, allowing for frontal observation of the boiling process. The active length of tubes was 150 mm.

After multiple tests carried out during construction and calibration of the heating unit, as well as litera-ture research, it was decided to realize the concept shown in Fig. 3b with two copper bushings, separated with a 4 mm wide Bakelite ring installed between a cartridge heater and inner tube surface. Thermocouples were laid in grooves (0.55 × 0.55 mm) carved (electric groove carving) in bushings, while their ends were fixed in reces-ses of Ebonite rings. The used thermocouples were of the type K, with jacket diameter of 0.5 mm. Custom-built cartridge heaters have an external diameter of 4 mm, effective heating length of 150 mm and maximum power output P = 300 W.

a) left view b) right view

17 mm(20 mm)

10 mm

180 mm

150 mm

17 mm(20 mm)

145

mm

A

A

A - A

4,0

mm

8,8m

m

10 m

m

4 mm150 mm

180 mm

21 3 4 5 6

a) Tube bundle diagram b) Individual tube diagram

Fig. 3. Diagram of the investigated tube bundle. 1 – Thermocouple, 2 – Ebonite ring, 3 – Cartridge heater, 4 – Copper bushing, 5 – Investiga-ted tube, 6 – Plug.


35

Determining heat transfer coefficient for individual tubes of a bundle required individual measurements of the supplied electric power. These measurements are made with N13 – MPS unit, a digital programmable instrument designed for three-phase power grid measurements in symmetrically and asymmetrically loaded systems. It simultaneously displays measured values, digitally transmits them and converts into standardized analogue signal. Power indications take into account programmed ratio values. Each measured value is trans-mitted to a supervising system via RS-485 interface and then through a converter and RS-232 connection to a PC unit. The measurement unit can be operated from the PC screen through a series link, using dedicated software (WizPar).

The selected measurement unit has a nominal input current I = 5 A, phase input voltage Un = 3 × 230 V and programmable analogue current output -20 to +20 mA and interface RS-485. It is a standard instrument with quality control attestations.

Smooth power control of each individual heater is provided by thyristor-based precision power con-trollers, RPG1-RPG20. In addition there are Reg1-Reg20 regulator units which protect cartridge heaters from excessive temperature increase and prevent exceeding a certain temperature in the working agent’s circuit. Each regulator unit is connected with one of the thermocouples installed at a heater controlled by this regulator. Connection, disconnection and alarm temperature levels were programmed into the regulator units. Each of them is equipped with two displays which show the setpoint and a real measured value, so they are also used as temperature indicators. All regulating units Reg1-Reg20, just like the MPS measurement unit, were linked to a PC unit via a converter and RS-232 standard connection. This allows one to monitor their operation, change parameters and visualize temperature variations over time for any selected heater. Connection of the MPS unit also allows generating charts showing, e.g. load on individual phases or phase voltage values.

Measured momentary values of temperatures and pressure are transmitted to a computer – at a selected sampling rate – through an AL154RX02 multiplexer. The device has been designed for measurement duties at a laboratory setup. It is equipped with 127 measurement inputs for K-type thermocouples, pressure measure-ment input (-0.1 MPa to +0.6 MPa) via a custom-made pressure sensor AR26 (maximum measurement error 0.1% of measured value). The temperature measurement error is ± 0.05 K and the sampling time is set to 0.5 s. Data is transferred to a PC unit via a USB interface, while configuration programming is made with a dedicated APEK Multiplekser software which allows one to define the number of measurement channels, program input characteristics, calibration, visualization of all channels on bar graphs, visualization of selected channels on a chart, setting sampling time/rate, limiting measurement time and transferring data directly to Microsoft Excel files.

Fig. 4 shows a block diagram of the described power supply system coupled to the temperature measure-ment system and data acquisition system.

˜3 x 230 V

RS 232 / 485

RS 232 / 485

USB

PC


Power measurement unit

Power controllers

Fig. 4. Block diagram of the power sup-ply, temperature measurement and data acquisition systems.

36

The tube bundle was illuminated using a laser sheet system with a power of 1500 mW (532 nm, 10 kHz, multimode) with light sheet optics, installed perpendicularly to the investigated tubes. A CCD camera was in-stalled coaxially with tubes. It is based on the Texas TC 237 sensor and used to take photographs of the boiling process (up to 15,000 frames per second). The layout and connections of the imaging system and its real-life view are presented in Fig 5.

a) Diagram

The unit synchronizing operation of the CCD camera light sheet allows one to operate (turn on or flash) the laser when the camera is triggered. Also the flash length can be adjusted.

Supporting structures for the CCD camera and laser system are flexible so it is possible to adjust the illu-minated point and recorded spot in two planes – horizontal and vertical.

The laser lighting system is also equipped with a special optical system regulator allowing changing the laser sheet geometry (vertically-horizontally, beam width).

A detailed description of the experimental setup and measurement procedure has been published in [9].

3. INVESTIGATION RESULTS

Experiments revealed that regardless of spacing and pressure values, the highest heat transfer coefficient is observed for water – which results from its excellent thermophysical properties, above all the high heat of evaporation. For example, Fig 6 presents a boiling curve in a smooth tube bundle with s/d = 1.7, at atmospheric pressure.

b) Real-life view

Fig 5. Lighting and imaging system.


Lighting

Digital video camera

Photo camera

PC unit CCD camera Laser power supply with synchronizing

unit

Measurement tank with heating section

Laser

CCD camera

Adjustable support

Laser

Laser power supply unit

37

For all investigated liquids, both in atmospheric and subatmospheric pressures, higher heat transfer coef-ficients were observed for the larger of the two investigated spacing values, i.e. for s/d = 2.0. For example, Fig 7 presents a boiling curve for R141b refrigerant at atmospheric pressure.

Regardless of the boiling liquid type and spacing, higher heat transfer coefficients were observed at at-mospheric pressure, which complies with several published experimental reports. For example, Fig 8 presents a boiling curve for methanol boiling in a bundle of tubes with porous coating for s/d = 2.0.

Fig 9 shows a relation between the average heat transfer coefficient for a specific tube row, and investi-gated heat flux values for water boiling in a smooth tube bundle, s/d = 1.7, at atmospheric pressure. The higher the tube row is, the higher the heat transfer coefficient gets. The heat transfer coefficient value also increases along with the heat flux value. Such a distribution of the heat transfer coefficient values can be explained by heat transfer intensification caused by vapour bubbles generated on lower tube rows. Visualization of this phenome-non is shown in Fig 10.

8 10 12 14 16 18 20 22 2415

20

25

30

35

40

45

50

q [k

W/m

2 ]

T [K]

12 14 16 18 20 22 2415

20

25

30

35

40

45

50

q [k

W/m

2 ]T [K]

Fig 6. Impact of the boiling liquid type on the boiling curve at atmospheric pressure (p = 101.2 kPa), in a tube bundle with s/d = 1.7.

Fig 7. Impact of spacing on the boiling curve for R141b refri-gerant in a smooth tube bundle, at atmospheric pressure (p = 101.2 kPa). s/d values: + - 1.7, × - 2,0

10 12 14 16 18 2015

20

25

30

35

40

45

50

q [k

W/m

2 ]

T [K]

Fig 8. Impact of pressure on the methanol boiling curve for bundle of tubes with porous coating, s/d = 2.0+ - atmospheric pressure (p = 101.1 kPa) × - subatmo-spheric pressure (p = 19.5 kPa

Fig 9. Relation between the average heat transfer coefficient for different heat flux values and the location of a tube row within a bundle – water boiling in a tube bundle, s/d = 1.7, atmospheric pressure (p = 100.5 kPa).


No. of tube row within a bundle

38

Fig 11 presents variability of the bundle factor (F) and bundle effect (WP) for methanol boiling in a tube bundle with s/d = 2.0 at atmospheric pressure. In compliance with previously published data, the bundle effect factor is somewhat higher than the bundle intensification coefficient and both values decrease as the heat fluxgrows.

Due to the fact that each tube in a bundle is equipped with its own heater, it is possible to determine the distribution of a heat exchange coefficient within the bundle. For example Fig 12 presents approximate lines ofconstant heat exchange coefficient for water boiling in a smooth tube bundle at atmospheric pressure for mini-mum and maximum investigated heat flux values. Heat transfer coefficients in the bottom part of both bundlesare almost half those observed in the topmost rows. Additionally the values for the central column are some-what higher than to the sides.

15 20 25 30 35 40 45 501,8

2,0

2,2

2,4

2,6

2,8

3,0

WP,

F [-

]q [kW/m2]

Fig 10. Visualization of intensifying steam bubble effect for water boiling process in a tube bundle s/d = 1.7, at atmospheric pressure (p = 100.5 kPa) and heat flux q = 30.25 kW/m²

Fig 11. Bundle factor (F - +) and bundle effect (WP - o) for R141b refrigerant boiling in a smooth tube bundle, s/d = 2.0 and subatmos-pheric pressure.

Application of multidimensional regression analysis for the boiling process of water, methanol and R141b refrigerant, for subatmospheric pressure and atmospheric pressure, allowed developing a correlation equation (1) which can be used to determine the average value of heat transfer coefficient for smooth tubes.

(1)

a) b)

2,5 2,9 2,7

2,5 2,1 2,4 2,1

1,7 1,8 1,9 1,7

1,6 1,6 1,7 1,5

1,5

1,8

1,41,5

2,6

3,6 3,8 5,0

3,6 3,0 3,2 3,0

2,8 3,0 3,2 2,8

2,7 2,7 3,0

2,6

3,0

2,42,8

Fig 12. Lines of constant heat transfer coefficient for water boiling in a tubebundle s/d= 1,7a) q = 15,42 kW/m2, b) q = 49,17 kW/m2; Values given in [kW/(m2K)]

67,0

74,048,12

305,0 Prln7,521

Ds

ppBoNukr


39

where: Nusselt number

l

dNu

, boiling number

lv

l

rLq

Bo

,Prandtl number

a

Pr , charac-

teristic dimension

vlgL

a – thermal diffusivity, m²/sd – diameter, mg – apparent gravity, m/s2

p – pressure, N/m2

q – heat flux density, W/m2

r – heat of evaporation, J/kgs – tube spacing, mα – average heat transfer coefficient, W/m2Kλ – thermal conductivity, W/mKρ – density, kg/m³µ – dynamic viscosity, Ns/m² σ – surface tension, N/mv – kinematic viscosity, m²/skr – criticall – liquid v – vapour

Comparison of the average heat transfer coefficient observed in experiments with the values calculated with the proposed formula shows that only 4 out of 72 points do not fit the ± 20% range (Fig 13).

Fig 13.Comparison of experiment results with calculations for boiling of water, methanol and R141b refrigerant in a smooth tube bundle at atmospheric pressure and subatmospheric pressure.


Water

Methanol

R 141b

α pr

ed [

kW/m

2 K]

α exp [kW/m2K]

40

REFERENCES

4. CONCLUSION

The systematic experimental research on the boiling process of water, methanol and R141b refrigerant in a smooth tube bundle shows that:

• Regardless of tube spacing and pressure value, the highest heat transfer coefficients are observed for water

• Regardless of pressure and type of boiling liquid, higher heat transfer coefficients are observed for the larger of the two investigated spacing values, in the case of s/d = 2.0

• Regardless of the type of boiling liquid and tube spacing, higher heat transfer coefficients are observed for atmospheric pressure

• Proposed correlation for boiling of water, methanol and R141b refrigerant provides results convergent with those obtained during experiments.

1. Marto PJ., Anderson CL., Nucleate boiling characteristics of R-113 in small tube bundle. ASME J. Heat Transfer, vol. 114, p. 425-433, 1992.

2. Memory S.B., Chilman SV., Marto PJ., Nucleate pool boiling of TURBO-B bundle in R-113. ASME J. Heat Transfer, vol. 116, p. 670-678, 1994.

3. Qiu Y.H., Liu Z.H.: Boiling heat transfer of water on smooth tubes in a compact staggered tube bundle. Applied Ther-mal Engineering, vol. 24, p. 1431-1441, 2004.

4. Gupta A., Saini J.S., Varma H.K., Boiling heat transfer in small horizontal tube bundles at low cross-flow velocities. Int. Journal of Heat and Mass Transfer, vol. 38, no. 4, p. 599-605, 1995.

5. Kumar S., Mohanty B., Gupta S.C., Boiling heat transfer from vertical row of horizontal tubes. Int. Journal of Heat and Mass Transfer, vol. 45, p. 3857-3864, 2002.

6. Leong L.S., Cornwell K., Heat transfer coefficients in a reboiler tube bundle. The Chemical Engineer, 343, p. 219-221, 1979.

7. Gupta A., Enhancement of boiling heat transfer in a 3x5 tube bundle. Int. Journal of Heat and Mass Transfer, vol. 48, p. 3763-3772, 2005.

8. Webb R.L., Choi K.D., Apparao T.R., A theoretical model for prediction of the heat load in flooded refrigerant evapo-rator. ASHRAE Trans., vol. 95, Pt. 1, 326-338, 1989.

9. Krasowski K., Przejmowanie ciepła przy wrzeniu na poziomym pęku rur z powłoką porowatą. PhD thesis, Faculty of Mechanical Engineering, Gdańsk University of Technology, 2009.


42


Zdzisław KustoGdańsk / Poland

Zdzisław Kusto graduated in power plant and energy management from the Faculty of Electrical Engineering of Gdańsk Technical University. Professionally he is associated with the Faculty of Electrical and Control Engineering at Gdańsk University of Technology, Institute of Fluid Flow Machinery of The Polish Academy of Sciences in Gdańsk, and Gdańsk College of Administration. He specializes in the issues of efficient use of unconventional energy sources, including renewable energy sources (sun, wind, biogas, biomass, and heat pumps). He has directed, among others, postgraduate studies: Nuclear Plant Construction, Nuclear Plant Design, and Energy Auditing. He was a member of the Council of Experts of the National Agency for Energy Conservation (1998–2001), and since 1997 at Gdańsk University of Technology he has been a member of the Rector’s Committee for Open University. He co-founded the Polish Wind Energy Association and the Foundation for Energy Conservation in Gdańsk, of which he is a member. He is the author of, among others, two monogra-phs on the economic efficiency of heat pumps, and the author and a co-author of 140 scientific and engineering publications and papers, and of 10 reviews of books and research studies.

Zdzisław Kusto / Gdańsk University of Technology

43

INCREMENTAL COST METHOD IN COST-EFFICIENCY EVALUATION OF INVESTMENT IN DISTRIBUTED GENERATION


1. INTRODUCTION

Small energy sources, including unconventional sources, are categorised as distributed sources that sup-ply minor recipients. Often they must cooperate with conventional sources, thereby creating hybrid sources1.

Various investment economic efficiency evaluation methods have been employed, out of which the follo-wing discount methods deserve particular attention: the annual cost method [1, 2], which has been practised in the power sector for years, as well as NPV and IRR methods [3, 4].

The above-mentioned methods can be assessed as complete in theoretical terms. The NPV (net present value) method as applied to capital expense projects in the power sector takes into account the annual income from sale of heat and/or electricity, and annual costs associated with their generation. These current annual ba-lances of cash flows from subsequent years of a project’s operation are discounted to the year zero - preceding the year of the project’s commissioning.

As regards small distributed energy resources (fuel cell, photovoltaic plant, solar heating system, heat pump, etc.), which supply small individual recipients, usually no electricity and/or heat is sold, and the income can be interpreted as a reduction in the annual maintenance expenses of the conventional plant, mainly for the purchase of fuel and electricity. This leads to a modification of the classical NPV method. The cost of energy generated in a conventional plant is compared here as the reference cost with the generation cost of a new proposed source.

In the NPV method the aforementioned generation costs of energy, which constitute a component there-of, are discounted sums of annual current costs from a period of K years, where K = 1, 2,…. N (N – expected/assumed duration of the new source operation). Such conceived streams of annual costs for each period of K years are used to formulate the incremental cost method (IC), which in its form is very visual and the algorithm of which resembles the LLC method2.

The name “incremental cost (IC) method” has been proposed by this author. It illustrates the discount accumulation of expenses incurred in subsequent years of a proposed project procurement and operation.

2. INCREMENTAL COST METHOD COMPONENTS

The basis for the development of the IC method is classical principles of economic calculation, and it as-sumes the project user’s viewpoint in its cost calculation formula. A simplified graphic rendering of the method components is presented in Tab. 1. As in the NPV method, the following factors may be thereby considered:

• variation annual of revenues and operating costs,• variation of annual repayments of bank loans, incl. interest and commission,

Abstract

Economic efficiency of a small unconventional source of distributed generation is calculated by com-parison of the source’s costs of heat and/or electricity

generation with the cost of conventional generation. This paper describes the incremental cost (IC) method, which resembles the long-known LLC method.

1 In the hybrid regime of a co-generation source that generates heat and electricity, such as a thermal-electric power station, the same type of energy is generated by a number of sources (either heat or electricity).2 The LCC method was formulated in the USA. Its name is the acronym of Life Cycle Costs. It has been quite popular also in Poland.

Incremental Cost Method in Cost-Efficiency Evaluation of Investment in Distributed Generation

44

• variation of borrowing, discount, and inflation rates,• variation of interest on income.

Tab. 1. Illustration of IC method components

Current annual costs Annual balance discounted to the year zero

Annual Expenses discounted to the year zeroYear Revenues Expenses Annual balance

1 P1 W1 B1 = P1 – W1 BD1 WD1

2 P2 W2 B2 = P2 – W2 BD2 WD2

* * * * * *

* * * * * *

i Pi Wi Bi = Pi – Wi BDi WDi

* * * * * *

* * * * * *

N PN WN BN = PN – WN BDN WDN

BDi = Bi (l + d)–i WDi = Wi (l + d)–i

Annual balance stream discounted to the year zero SD = ∑ BDi

Annual balance stream discounted to the year zero SWD = ∑ WDi

SD NPV

SWD MKN

In individual years in the subject period of N years of the project’s operation the user will incur expen-ses. “Expenses” mean here fixed and variable operating costs3, bank loan repayment, income taxes, and loan servicing costs (interest, bank commission) and return of the capital expenditure portion covered with own contribution.

The annual expense sum discounted to the year zero (SWD) constitutes for a period of K years (K = 1, 2, ..., N) the incremental cost and is the basis for development of the modified annual cost method not describedhere4.

Because of calculation convenience the discounted expenses and revenues (revenues in the NPV method) are as a rule aggregated to the year zero. This is not an absolutely necessary condition. They may be aggregated to any year without compromising the calculation accuracy and precision.

The discounted sums (SD, SWD) are calculated using a discount factor – (1 + d)-i– (compound interest, j = 1, 2, ..., N), where – d is the discount rate, which usually takes a constant value in the calculation period of the operation – N years.

The annual cash flow in the jth year is the sum of all expenses – Wj, which is called the new plant’s current costs – Wj =Krhj.

i–N

i=1

i–N

i=1

3 In the power sector the term “variable operating costs” means: the costs of energy and fuels purchased for the electricity and heat generation, and the costs of so-called operating supplies. The other operating cost items (overhaul, current repair, administration, etc.) that are independent of the generated output are called „fixed operating costs”.4 The annual cost method was developed in the 1960s by Prof. K. Kopecki with the intention to employ it in evaluation of cost-efficiencies in the powersector. In its overall assumption the method is general and may be employed in various branches of the economy. It has been applied in the power sector until now and is particularly relevant to selection of the optimal investment option.


45

Wj = Krhj = Kr ni j + Krk j + Pdochj / zł/a (1)

• new unconventional plant

Kr ni j = Kest ni j + Kezm ni j + Zkr ni j + Pkr ni j + Amw ni j + Kdod ni j / zł/a (2)

• conventional element of hybrid plant (where hybrid source is necessary)

Krk j = Kest k j + Kezm k j + Zkr k j + Pkr k j + Amw k j + Kdod k j / zł/a (3)

j = 1, 2, 3, ..., N

where:K r mj – annual current costs of the new plant in the jth yearK rkj – current costs of the interoperable conventional plant in the jth yearKest x j – fixed operating costs in the jth year, inclusive of personnel, overhaul, and current repairK ezm x j – variable operating costs in the jth yearZkrj, – loan repayment instalment in the jth year (Zkryj = 0 where a bank loan was not contracted or is re-

paid) P kr x j – tax on the unpaid part of the loan in the year together with banking services (commission) (P =

0 where a bank loan was not contracted or is repaid)Pdochj – tax on income from the sale of energy generated in the hybrid /co-generation system in the jth

yearAmwxj – annual instalment of return on own contribution to the investment in the jth yearKdodaj – additional costs, if any, incurred in the jth year of the project’s operationPdochj – tax on income from the sale of heat in the jth year of the project’s operation.

Indexes x:ni – for the new plant

k – for the conventional plant interoperable with the new plant in the hybrid regime.

Total expenditure of funds in the year zero is the investment expenditure incurred, which may consist of own contribution of the proposed plant’s future user and of a bank loan. It may happen that the user receives a grant for the project, owing to which it may feel an (apparent) reduction in the investment expenditure.

B0 = (Kinwc - Dot) = [(Kinww + Kinwb) - Dot] / zł (4)

where:Kinwc – total investment expenditure on the project balanced to the year zeroKinww – user’s own contribution to the investment expenditure on the project balanced to the year zero Kinwb – bank loan to cover the investment expenditure on the project balanced to the year zero Dot – investment grant.

The total own contribution – Kinww and the loan – Kinwb may be divided between the new plant and the conventional plant interoperable with the new one in hybrid regime.

Kinww = Kinwwni + Kinwwk / zł (5)

Kinwb = Kinwbni + Kinwbk / zł (6)


46

Investment grantThe grant can take different forms, but it may ultimately be presented as a one-off amount lodged in

the year zero. The problem of subsidies and justification for their amounts has been repeatedly discussed bythis author in assessing the costs of electricity generation in wind turbines (e.g. [5–7]). One of the arguments presented to justify a subsidy for renewable and unconventional energy sources are so-called external costs associated with conventional power generation, which were calculated after many years of research conducted in the last decades of the previous century by an international team of experts Extern-E5 [e.g.: 8 –10]. Avoidance of at least some of these additional costs might be grounds for granting a subsidy.

A particular form of investment grant may be the use of effects of international or inter-regional coopera-tion between countries/ regions in which the investment expenditures on a project (solar heating system, heat pump, wind farm, etc.), as well as the generation/manufacture costs of energy or other products, vary. This problem has been long known to economists, and it was also analysed by this author with regard to wind farms with showing a very high potential efficiency of such co-operation clearly beneficial for the both parties concer-ned [11]. This concept, addressed to wind generation, may also be adopted for other distributed sources.

Investment expendituresCapital expenditures for a new plant are usually incurred in a period not longer than one year, during the

year zero.If the investment expenditures are incurred over several years, as is the case with a plant with high in-

stalled power (at least a few megawatts), their individual components from formula (4) will be the sums disco-unted to the year zero. Where the discount rate is fixed, the total expenditure may be calculated according to(7). Where the discount rate is variable over time (a generalized case), the total investment costs (K inwcd) are calculated after formula (8).

(7)

(8)

where:

Kinwc j – total investment expenditures incurred on the plant in the jth year, j = 1, 2, ..., N / PLN / adk – discount rate in the kth year, k = 1, 2, ..., NN1 – number of years of the investment expenditure’s incurrence, prior to the year of the project’s com-

missioningN2 – number of years of the investment expenditure’s incurrence, after the year of the project’s commis-

sioning.

In a similar way the discount sums of the total expenditure’s components may be calculated. The total expenditure’s components include the future project user’s own contribution and bank loans. The following formulas are presented in their generalised form, which takes into account the investment expenditures’ incur-rence in the years prior the year zero and in the year zero (prior to the project’s commissioning), and during the project’s operation (so-called phased capex project).

0

1

2

1)1()1(

Nj

N

j

jjinwc

jjinwcinwcd dKdKK

2

1

1

0

1 1 )1()1(

N

jj

kk

jinwc

Nj

j

kkjinwcinwcd

d

KdKK

5 Extern-E’s research in the 1980s and 1990s focused on calculation of the external costs of electricity generation. In the early 2000s research was planned on the external costs of heat generation.

/ zł

/ zł


47

Own contribution

(9)

where:Kinww j – own investment expenditures incurred in the jth year, j = 1, 2, ..., N / PLN /

A bank loans

(10)

where:Kinwb j – bank loan for the plant project drawn in the jth year, j = 1, 2, ..., N / PLN / a

Return of own investment contribution Also return of own contribution to the investment should be included in the cost of electricity and/or heat

generation. The heating system user does not spend this portion of money, but must accumulate it for reimbur-sement of its own investment expenditures.

The own contribution is returned during the project’s operation. The annual cost method used in the second half of the last century assumed that it lasts throughout the project’s calculation lifetime – i.e. for N years (N am = N). Currently, the amortization period is typically assumed shorter than the calculation period: Nam < N.

The own contribution must be fully recovered, which means that the sum of all annual repayments in Nam years - discounted to the year zero6 must be equal to the stream of own contribution to the investment also discounted to the year zero.

(11)

The return of own contribution to the investment (Amw j) in formula (11) may vary from year to year. Pro-vided that it is a constant value.

Amw j = Amw = const

then dependency (11) takes form (12):

(12)

2

1

1

0

1 1 )1()1(

N

jj

kk

jinww

Nj

j

kkjinwwinwwd

d

KdKK

2

1

1

0

1 1 )1()1(

N

jj

kk

jinwb

Nj

j

kkjinwbinwbd

d

KdKK

/ zł

/ zł

Nam

jj

kk

jmwmort

d

AA

1

1

)1(

inwwd

am

mwmort K

rA

A

6 As has already been mentioned in this chapter, all cash flows may be discounted to any year, but the year zero is the most convenient.

/ zł

/ zł/a


48

where:ram – capital return rate calculated for the period of Nam years

If dam = 0 (!), then:

dam – average discount rate in the period of Nam years, which in a general case may be calculated as the geometric average.

Tax on income from the sale of heat and electricityIn large systems (e.g. housing project heating) energy is sold. Any such heat generator must pay tax on

the income from the sale of heat in the subsequent years of the period of N years.

Pdoch j ≠ 0, j = 1, 2, 3, ..., N

For low-power heating systems that supply individual customers, the heat output is not sold, but consu-med directly by the recipient. In this case, the tax is equal to zero.

Pdoch j = 0, j = 1, 2, 3, ..., N

Operating costsThere are two kinds of operating costs. Variable operating costs are the costs of energy and operating

supplies consumed in subsequent years of the N-year period. For a new plant that may be calculated after for-mula (13), for a conventional plant – after equation (14), considering that these costs may vary in subsequent years due to changing electricity and fuel prices.

• for hybrid sources with heat pumps

(13)

(14)

• for other hybrid source types

(15)

(16)

amam Nr 1

1)1(1

1

Nam

Nam

ttamam dDd

1)1()1(

Nam

am

Namamam

am ddd

r

elruch

sil

ipcpcjmrpcjeljniezm E

TQkcK

kd

ikikjmrkjpalkajmrkjpaljezmk W

TQkcBkcK

aeljnimrjeljniezm EkcK

kd

ikikjmrkjpalkajmrkjpaljezmk W

TQkcBkcK

/ zł/a

/ zł/a

/ zł / a

/ zł/a


49

where:cel j – expected electricity price in the jth year, PLN/kWhcpalj – expected fuel price in the jth year, PLN/kg7

Eelruch – annual consumption of electricity for operating purposes, kWh/akmk ni j – operating supplies cost coefficient in the jth yearkmrk j – operating supplies cost coefficient for conventional plant in the jth yearkmr pc j – operating supplies cost coefficient for heat pump in the jth yearQik – installed power of the boiler, kWQ ni – installed power of the new plant, kWQpc – installed power of the heat pump, kWTini – installed power use duration of the new plant, h/aTik – installed power use duration of the boiler, h/aTipc – installed power use duration of the heat pump, h/aWd – calorific value of the fuel, kJ/kg or kJ/m3 8

φ – coefficient of performance of the heat pumpηk – boiler efficiency, annual averageηsil – efficiency of the motor driven by the heat pump.

Fixed operating costs may be calculated as fixed amounts in subsequent years, using the fixed operatingcosts rate [1]. It must be clearly underlined that these costs may change in subsequent years. Anticipating these changes is a forecasting issue.

(17)

(18)

Bank loan repayment and service chargesBank loan (Kinwb) is repaid throughout the period of Nb years. It is assumed that the annual loan repayment

instalments (Zkrj) are constant and equal.

(19)

The loan accrues interest at the interest rate – pkr, which determines the amount of interest in the jth – Qpr j year, calculated on the unpaid part of the loan.

(20)

7 The fuel price may refer to a unit of fuel mass (hard coal, lignite, oil and oil derivates – 1 kg) or to a volume unit of gas fuel (1 m3). This price may also be quoted for a unit of the fuel’s calorific value. For thermal oil: 1 kg = 42 MJ = 0,042 GJ, for natural gas: 1 m3 = 35 MJ = 0,035 GJ.8 Calorific value for a solid fuel (hard coal, lignite, coke, etc.) is quoted in kilojoules (kJ) or megajoules (MJ) per one kilogram of fuel, and for a gas fuel(natural gas, distillation gas, municipal gas, etc.) it is quoted in kilojoules (kJ) or megajoules (MJ) per one cubic metre of fuel.

2

1

N

Njjniinwceniestjniest KrKK

2

1

N

Njjinwkceestkestkj KrKK

/ zł/a

/ zł/a

b

inwbjkr N

KZ

krkrinwbjpr pZjKO ))1(( / zł/a


50

Bank commission is calculated on the amount transferred to the bank at a set rate of commission – Pprb.

(21)

Additional costsThis item contains all additional and not yet specified cash expenses on the plant operations, e.g. envi-

ronmental costs directly arising from the plant operations and unexpected random costs not specified as a fixedoperating cost.

If resulting from operation of a heat system with a heat pump its user saves money, then, in accordance with the applicable the user is obliged to pay tax on the savings, which may also be a component of the additio-nal costs.

3. INCREMENTAL COST (IC) METHOD

As mentioned above, the incremental cost method is part of the NPV method. The IC method calculates the stream of annual current costs (SKLh) discounted to the year zero (!) for a hybrid plant for the period of L years, while the L number is counted incrementally, successively, up to the value of N. The incremental costs for the whole plant are the sum of individual incremental costs of the new facility’s systems and the incremental cost of the conventional plant.

(22)

The incremental cost method is particularly useful in comparing investment options. It has the following major advantages:

• it is a discount method, it is theoretically complete,• it provides a clear visual representation of all costs - investment expenditures and current costs. The method’s disadvantage is its computational complexity that requires good preparation, and it is

difficult to verify forecasting of numerical data (e.g. discount rates, changes in fuel and electricity prices, etc.),as in the annual cost (NPV) method or other discounting approaches. The IC method’s graphic presentation is demonstratively shown in Fig. 1, which illustrates the comparison of three investment options.

Under option 1 the initial costs are small in the first year of operation, which may originate from the rela-tively low investment expenditure, but the current costs are high. In period L1 the aggregate costs (discounted streams) are the lowest, but after a period of L2 they are already the highest. Option 2 features very high initial costs (high investment expenditure), but also the lowest annual current costs and already after L2 is better than option 1, and after period L3 it is the cheapest option. It may be noted that if an investment project’s time horizon is shorter than L1, then option 1 is absolutely the most advantageous, but if this period is longer than L3, then option 1 should be implemented.

prbkrjprjprow pZOK )(

L

jj

tkt

jrkL

jj

tpct

jrpcKLh

p

K

p

KS

1

1

1

1

)1()1(

/ zł/a

/ zł


51

Cost-effectiveness comparison of several investment options imposes the condition of meeting the basic requirement of equality of the final results for all options. With respect to plant variants the final results’ equality boils down to: the same annual energy output, the same peak heat/electric power and the variation thereof in time (equality of the time-ordered charts). Failure to meet the above requirements makes the options compari-son difficult. Where the final results differ, they have to be additionally equalised by computation.

All known calculation methods require bringing all investment options to the same final results.When the effectiveness of the investment in a project is assessed from the project user’s viewpoint, then

the initial costs include only those expenses that are directly incurred by the user, but no bank loan. During a heating system’s operation its user repays the whole debt, and this is included in the current costs of the heat and/or electricity generation, which also include the return of own investment contribution deducted in subse-quent years. The discussed calculations include all cost items, therefore they are substantially complete.

The incremental cost method can distinguish a special case, whereby the investment expenditure alloca-ted to the year zero is reassigned to the first year of operation, which theoretically means the repayment of all bank loan instalments (if granted) and return of the own investment contribution within this first year.

Fig. 1. Graphic illustration of the (annual) incremental cost method


Incr

emen

tal c

osts

Option 1

Option 2

Option 3

Incremental periods of L years

52

REFERENCES

1. Kopecki K., Materiały i Studia. Volume V. Zasady ekonomicznego rachunku [Principles of Economic Calculation], Part I, Ogólne założenia i metodyka rachunku gospodarczego w pracach planowo-projektowych w elektroenergetyce [General Assumptions and Methodology of Economic Calculation in Planning and Design Works in the Power Sector], The Polish Academy of Sciences, The Committee for Electrification of Poland, Warsaw 1960.

2. Bojarski W., Podstawy metodyczne oceny efektywności w systemach energetycznych [Methodological Grounds for Effectiveness Evaluation in Power Systems], The Polish Academy of Sciences, The Committee for Power Sector Problems, Wroclaw - Warsaw - Krakow - Gdańsk, The Ossoliński National Institute, The Polish Academy of Sciences Publishers, 1979.

3. Ratajczak E., Elektroenergetyka polska w okresie przemian [The Polish Power Sector in the Time of Transition], Gdańsk University of Technology, Faculty of Electrical Engineering, Gdańsk, 21-22 January 1993.

4. Górzyński J., Audyting Energetyczny [Energy Audit], The Polish National Energy Conservation Agency, Warsaw 2002.5. Soliński J., Solińska M., Ekologiczne podstawy systemu wspierania rozwoju energii odnawialnej w Polsce [Ecological

Grounds for the System of Renewable Energy Development in Poland], International Seminar “Wind Power Generation on Land and in Sea”, Sopot, 15-17 December 2000.

6. Kusto Z., Ekonomiczne, społeczne i ekologiczne warunki urynkowienia elektrowni wiatrowej [Economic, Social, and Ecological Conditions of Wind Power Plant Commercialisation], Scientific Symposium “Planning and Operation of Energy Supply Systems”, Gdańsk, 29-30 March 2001.

7. Kusto Z., Warunki rynkowego użytkowania elektrowni wiatrowych w nadmorskich miejscowościach Wybrzeża Gdań-skiego [Market Conditions for the Operation of Wind Power Plants in the Coastal Towns of the Gdańsk Coast], Scientific--Technical Symposium “Technical, Ecological, and Economic Aspects of Renewable Energy Generation”, Faculty of Produc-tion Engineering of Warsaw University of Life Sciences, Warsaw, 19-20 October 2001.

8. New research reveals the real costs of electricity in Europe, European Research Area, Brussels, 20 July 2001, http:// europa.eu.int./comm/research/press/2001/pr/200/en.html 01-07-31.

9. Malko J., Internalizacja kosztów zewnętrznych, czyli ile naprawdę kosztuje energia [External Costs Internalisation, i.e. How Much Does Energy Really Cos?t], Wokół Energetyki, October 2004.

10. Radovic U., Promocja wytwarzania energii elektrycznej ze źródeł odnawialnych w Polsce: czy dodatkowy koszt sys-temowy jest uzasadniony? [Promotion of Electricity Generation from Renewable Sources in Poland. Is Additional System Cost Justified?] Polityka Energetyczna, Vol. 8, Special Issue, 2005, PL ISSN 1429-6675.

11. Kusto Z., Wpływ efektów współpracy międzynarodowej na nakłady inwestycyjne na elektrownię wiatrową [Interna-tional Cooperation Impact on Investment Expenditures on Wind Power Plant], VIII National Forum of Renewable Energy Sources, Międzybrodzie Żywieckie 15-17 May 2002 and Warsaw, 28-30 October 2002.


53

SUPPLEMENTARY REFERENCES

1. Berent-Kowalska G., Kasprowska J., Kasperczyk G., et. al., Energia ze źródeł odnawialnych w 2006 roku [Energy from Renewable Sources in 2006], The Central Statistical Office, Department of Industry, Ministry of Economy, Department of Power Sector, Warsaw 2007

2. Joosen S., Wahlström Å., Sijanec Zavrl M., Makowska N. et. al.: Studium wykonalności dla alternatywnych systemów energetycznych [Feasibility Study for Alternative Power Systems], Czysta Energia, January 2009.

3. Kamiński S., Zadania sektora paliwowo-energetycznego w zakresie środowiska w świetle integracji z Unią Europej-ską [Environmental Tasks of the Fuel and Energy Sector in View of Integration with the EU], Conference „Energy policy of Poland in the Years to Come”, Warsaw, 6-7 March 2002.

4. Kulesa M., Planowanie energetyczne w gminie, Generacja rozproszona (kogeneracja gazowa, źródła odnawialne) oraz przedsiębiorstwa multienergetyczne w strategii gmin. Wybrane przykłady [Energy Planning in Municipality, Distri-buted Generation (Gas Co-generation, Renewable Sources) and Multi-energy Companies in Municipal Strategy. Selected Examples] Energetyka, January 2003.

5. Public Notice of the Minister of Environment dated 20 September 2007 on the environmental charge rates in 2008, Monitor Polski # 68, Item 754.

6. Regulation of the Minister of Economy and Labour dated 9 December 2004 on the detailed scope of the obligation to purchase electricity generated in renewable sources. Dziennik Ustaw # 267, Items 2655 and 2656.

7. Regulation of the Minister of Economy and Labour dated 9 December 2004 on the detailed scope of the obligation to purchase electricity co-generated with heat, Dziennik Ustaw # 267, Item 2657

8. Strategy for the Development of Renewable Energy Sector, Ministry of Environment, Warsaw, September 20009. Szramka R., Rozwój i regulacja rynku energii odnawialnej w Polsce [Development and Regulation of the Renewable

Energy Market in Poland], Biuletyn URE 5/2003.10. Strategia rozwoju w Polsce wysokosprawnej kogeneracji - główne kierunki [Strategy for the Development in Poland

of High-Efficiency Cogeneration – Main Directions], Study developed under the direction of Prof. Janusz Lewandowski, Con-tract 501H/4433/0445/000, Power and Environmental Protection Research Centre of Warsaw University of Technology, Institute of Thermal Technology of the Silesian University of Technology, Warsaw, June 2007

11. Świderski M., Analiza LCC (Life Cycle Cost Analysis) narzędziem wspomagającym ocenę projektów inwestycyjnych związanych z techniką pompową [LCC (Life Cycle Cost Analysis as Supporting Tool for Pump Technology-Related Invest-ment Project Evaluation], IX FORUM OF PUMP USERS, Szczyrk, 1-3 October 2003.

12. Palka-Wyżykowska K., Metoda LCC i jej przydatność do ekonomicznej oceny efektywności systemów energetycz-nych na przykładzie systemów grzewczych w budownictwie mieszkaniowym, Opracowanie SiUChKl, [LCC Methods and its Usefulness in Economic Evaluation of Power System Efficiency in Example of Heating System in Residential Construction, SiUChKl Study] Faculty of Mechanical Engineering of Gdańsk University of Technology, Gdańsk 2008.


54


Zbigniew LubośnyGdańsk / Poland

Zbigniew Lubośny graduated from the Faculty of Electrical and Control Engineering at Gdańsk University of Technology in 1985. In 1991 he defended his PhD thesis, and eight years later he obtained the assistant professor degree at the same university. A professor of engineering since 2004. Now works at Gdańsk University of Technol-ogy as an associate professor. His area of research includes mathematical modelling, power system stability, power system control, artificial intelli-gence application in power system control, wind power plant modelling and control.

Zbigniew Lubośny / Gdańsk University of Technology

55

MAXIMISATION VS. OPTIMISATION OF RADIAL POWER DISTRIBUTION GRID USE


1. INTRODUCTION

Power system development is conditioned by significant, if not massive, increases in the number of elec-tricity sources of various types, i.e. distributed generation, connected to power grids. The sources are connected to high voltage (HV), medium voltage (MV), as well as low voltage (LV) grids. Mainly large wind farms and larger gas, biogas, and biomass power plants are (and will be) connected to HV grids. To MV grids, smaller (in terms of the rated power) plants of the aforementioned types are (and will be connected). In LV grids, meanwhile, mainly home and household micro-power plants can be expected, including wind, photovoltaic and thermal plants fired with typical fuels such as coal, gas, an liquid fuels.

In the past, power systems evolved in the direction of systems in which resistance to changes in load regimes, i.e. power received, was a basic requirement. This condition still exists in practice.

At the same time it was, and still is, required that a change in the configuration of a grid to which loads are connected, meant as outage of a system item (and in some cases of two system elements) leads to neither overloading of the other grid items nor the voltage’s deviation beyond the allowable range.

The above means that such (contemporary) grids operate with a large transmission capacity margin. The capacity margin must be understood here in the long-term context, i.e. in a time horizon of at least 24 hours, and in practice throughout a year or longer. Because in the short term, e.g. at a peak load, it may happen that the transmission capacities of a grid or its parts (components) are substantially (or fully) utilized.

Power grids which operate in the manner described above, have been and still are developed based on the principle that might be called connect and forget. This means that to such a grid, loads are connected, the operations of which at any point in time, in normal conditions, can cause no excess of the grid items’ allowable loads.

Expansion of so-called distributed generation entails connection of various types of energy sources to distribution grids. From the grid operator viewpoint electricity sources may be connected to a grid according to the following principles:

1. Connect & Forget. Sources are connected to a grid, the operation of which at their respective maximum (rated) powers leads neither to the grid’s overloading, nor excesses in the voltage level. The system operator in this case is compelled (by technical reasons) to neither keep track of the generated power, nor to control the actual loads of the grid items (power lines, transformers) and the voltage levels.

A power grid operated according to this principle should be regarded as the least utilised in terms of its transmission capacity.

2. Connect, Control Locally, and Forget. In this case, an energy source must be provided with a regula-tor/limiter of its active power output to the grid. The regulator/limiter setpoint is meant to preclude the power grid’s overloading. Such source’s control system operates autonomously, using only local measurements. The system operator here does not adjust the source’s power output on-line, i.e. the source is not subject to opera-tor control.

Abstract

The article discusses issues of development of distribution grids saturated with so-called distributed ge-neration, in the context of the maximum and optimum use of these grids. The maximum use of a radial distribution grid is meant as the grid’s operation at the power lines’ loads equal or similar to those permanently allowed.

The optimum use of such a grid means its operating condition leading to minimising power and energy losses in the lines. Either of the foregoing operating conditions may be obtained by appropriate allocation of the distribu-ted generation in a grid of this type.

Maximisation vs. Optimisation of Radial Power Distribution Grid Use

56

A power grid operated according to this principle should be regarded as fairly well utilised in terms of its transmission capacity.

3. Connect and Control Actively. In this case, a source requires active control by the system operator. This follows from the source’s maximum (rated) power in relation to the grid structure and specification, includingits transmission capacity. This control can be executed in various ways, namely on-line or off-line. The source performance, including its output in various periods of time, may result from the system service provision by the source to the system operator.

A power grid operated according to this principle should be regarded as potentially capable of good use in terms of its transmission capacity. The extent of utilization of these grids’ transmission capacities is determined by the location and type of sources subject to the operator control, and the control and regulation algorithms applied.

Active influence on electricity source is a typical activity so far performed by the system operator only. Tofulfil this task the system operator is provided with the appropriate technical and organisational infrastructure.Subject to this control are not only high power sources, i.e. relatively few objects. This implies the relative ease of such process implementation in terms of IT infrastructure and control algorithms.

Increased saturation of power systems with small and very small electricity sources will necessitate in-volving a large number in regulation processes implemented in the power system. These processes include development of control infrastructure and systems (IT networks, measurement systems, regulation algorithms, operator’s organisational structures, etc.), so-called smart grids.

It also seems that a portion of these small sources will never be subject to centralised control. These sour-ces will be still connected according to the Connect & Forget or Connect, Control Locally & Forget principles.

2. MAXIMUM POWER OF SOURCES CONNECTED TO RADIAL GRID

2.1. Source connection at the end of branchBy far the majority of the low and medium voltage grids in the KSE National Power System operate as

radial grids. These are erected as overhead grids with bare or insulated wires, and as cable grids. A grid of this type may be supplied from a variety of electricity micro-sources, such as wind, water, and thermal (gas, biogas, and biomass) plants, or photovoltaic sources.

A typical grid structure is presented on Fig. 1. In such a grid the electricity sources may be connected to the low voltage (LV or MV) buses of a transformer that powers the grid (generator G1) and to the grid’s existing (receiving) or new nodes (generators Gj and GN).

G G G

N1 i32 j

nNSN

Fig. 1. Typical radial grid structure

WN (SN) SN (nN)


57

The maximum power output of a source, which, according to the Connect & Forget principle, may be connected to a radial grid such as that in Fig. 1, whereby the source is situated at the end of a radius and there are no other sources in the given branch equals1:

SnG,N = Sdd,N,N-1 + SmO,N (1)

where Sdd,N,N-1 is the maximum permissible load of a power line’s stretch N÷N-1 and follows directly from the line’s permanent current-carrying capacity, while SmO,N is the minimum power (over any period of time) re-ceived by the load in node N.

If in a given branch there already are other sources and the source closest to node N is connected to node j, the maximum power that can be connected to node N equals:

S(nG,N) = S(dd,i,j) + ∑l=j SmO,l – SnG,j (2)

This means that when SnG,j = Sdd,i,j, and the minimum loads SmO,l equal zero, then no energy sources may be connected to node N.

For the rated source output defined by (1) and (2) the power consumed by the load in node N, as well asthe source output power, are irrelevant from the permitted grid load-carrying capacity viewpoint. This means that even in the case of a source operated at its rated output and no power (energy) input to node N, the po-wer line will not be overloaded. This is therefore a connection of a source to a grid according to the Connect & Forget principle.

The following advantages of this arrangement can be identified:• Improved voltage conditions, i.e. reduced voltage drops and increased voltage stability margin (altho-

ugh it is not usually a problem in LV low voltage grids and to a lesser extent in MV grids),• No requirement for grid extension• Reduced power (energy) losses in the grid• No requirement for the grid operator’s involvement in the active management of grid operation (Con-

nect & Forget principle).However, an adverse element may be deteriorated energy quality. This will however depend on the energy

source characteristics rather than on the source’s connecting node or its rated power output. However, a certain measure of energy quality deterioration, i.e. to the level determined by standards and regulations, is permit-ted.

Increased use of the existing power grid is possible in the case of sizing a source to the power output equal to:

Sdd,N,N-1 + SmO,N < SnG,N ≤ Sdd,N,N-1 + SMO,N (3)

where S MO, N is the maximum power consumed by the load in node N. This power may in practice result from the rated current of the fuse in the service line of the load in node N.

In the above case, however, it is necessary to reduce power S (t) G, N generated by the source at a given moment of time t to the value:

S(t)G,N = Sdd,N,N-1 – S(t)O,N ≤ SnG,N (4)

1 In the discussions and dependencies presented here the energy losses in grid are neglected. Therefore the dependencies are approximate. The maximum (rated) source outputs are denominated as apparent powers. With some approximation they may, however, be treated as active powers. The above is a simplification aimed at the formulation of as simple as possible engineering/utilitarian dependencies.

N


58

A source of this type requires, therefore, a regulator/limiter of its power output to the grid. The regulator should measure the power or current over distance N÷N -1 and should maintain the difference between the powers generated P (t)G, N, and consumed in the node S (t)O,N, below, or equal to, the maximum power Sdd, N, N -1 or the current limit of that section of the line Idd, N, N-1. Maintaining the power equal to Sdd, N, N-1 (or current I,dd, N, N-1) im-plies the full utilization of the existing section of the grid, i.e. segment N÷N-1. The energy input to the grid will then be consumed by subsequent loads in the branch, i.e. the loads situated in nodes N -1 , or N -2, etc.

A source will be fully utilised when it is capable of electricity generation in times of maximum power consumption by loads in node N. Then the source will supply the loads in node N, and will additionally output electricity to the grid. At times when the demand for electricity in node N is below the maximum, the source will not be fully utilised.

Source operation at the power output that makes the power flow in the grid’s N÷N-1 section equal to the permitted maximum leads to increased power (energy) losses in this and – possibly – in one or several subse-quent sections, and to decreased losses in following sections. The overall effect is always the result of a specificgrid configuration.

In this case, compared to the previous one, the above mentioned advantages and disadvantages are re-tained.

The requirement appears here, however, to implement a source output control system in the form of a limiter of the power transmitted in branch N÷N -1.

2.2. Source connection to internal node of branch In the case of connecting to a radial grid, like the one on Fig. 1, of a source situated in node j, where

1 < j < N, i.e. situated between the first node 1 and the last node N, the sources’ maximum power output equals:• when beyond node j, i.e. between j+1 and N no energy source is installed:

SnG,j = min Sdd,i,j + SmO,j, Sdd,j (5)

• when downstream of node j, i.e. between j+1 and N, no energy source is installed:

SnG,j = min Sdd,j, Sdd,i,j + ∑k=j+1 SnG,k +∑k=j SmO,k (6)

where:Sdd,i,j – the maximum permitted power in line section i-jSdd,j – the maximum permitted power in the service line (or, possibly, an internal power supply line) Typically the service line’s transmission capacity is below that of the radial line, i.e. in

general dependence S d d , j ≤ Sdd,i-j is true.Sm0,k – the minimum load power (consumed power) in node k,S nG,k – the rated power of the source in node k.

Dependencies (5) and (6) are conservative, but consistent with the Connect & Forget principle. In either case, even the lack of power consumption in a grid branch will not result in an excess of the branch’s permitted load capacity.

It is worth noting that when the minimum load powers are equal to zero, and the rated output of a gene-rator, or the sum of the rated outputs of generators, situated downstream of node j is equal to the power limit of section j, and where the permanent load limits of all grid sections are identical, then no energy source may be connected to node j. In such a case grid section i-j will be loaded to the power limit anyway, i.e. with no source in node j.

In this case, compared to the previous one, the above mentioned advantages and disadvantages are re-tained.

As in the case of connecting a source to the end of a branch, the application of a system of regulation of the power input through a node (meant as the difference between the powers generated and consumed in the

N N


59

node) enables connection of sources with the power outputs in excess of that resulting from (5) and (6). It is a connection according to the Connect, Control Locally, and Forget principle. In this case, regardless of whether downstream of node j, i.e. between nodes j+1 and N, energy sources are or are not installed, the maximum so-urce power output should meet the condition:

SnG,j ≤ min Sdd,i,j + ∑k=j SMO,k – ∑k=j+1 SnG,k ,Sdd,j + SMO,j (7)

where the following condition should be satisfied:

Sdd,i,j + ∑k=j SMO,k – ∑k=j+1 SnG,k ≤ Sdd,j,j+1 (8)

In this case, the output controller of a power plant connected to node j has to maintain the set power below, or equal to, the transmission capacity of grid section i-j, or service line j.

In such a grid conditions may occur whereby the operation of sources closer to node 1 will be limited to the power consumed by loads in the source connection node. This may be, for example, if the whole branch (1 ÷ N) is made of one type of wire, and the loads do not receive power. Then the output to the grid of the last source in a branch will equal the branch’s power limit, and the other sources (provided that power and energy losses are ignored) will operate with the output equal to the power consumed by loads in the loads’ connection node.

This applies to the situation of connecting sources in the sequence from the branch’s end (node N) to its beginning. If a source situated in the middle of a branch has been connected first, the source will have (mayhave) the highest rated output. Then the other sources in the branch will have low output, or, if the minimum branch load (aggregate minimum power of loads in the branch) is equal to zero, they may not be connected. The connection sequence of the sources in the branch is of decisive importance here.

Connection of a source according to the specified principles still retains the foregoing advantages and disa-dvantages. However, a problem may arise involving the source output control system implementation. This applies to overhead grids - typical of villages and some single-family housing estates. Where there is a need for control-ling/limiting the power in branch i-j, instrument transformers should be installed on low voltage overhead line poles. This would complicate the control system and add costs. But this is not a technically unfeasible project.

2.3. Source connection to MV (LV) switchgear busesIf you connect an electricity source to MV (LV) switchgear buses (node 1 in Fig. 1), the maximum source

output equals:• in the absence of a source output control system, which conditions the output on the power flow in the

transformer: – where no other energy sources are connected to the grid:

SnG,1 = SnT + ∑l=1 (∑k=2 SmO,l,k) (9)

- where other energy sources are connected to the grid:

SnG,1 = SnT + kj × ∑l=1 (∑k=1 SmO,l,k – ∑l=1 (∑k=2 SnG,l,k) (10)

whereNg – number of branches in the radial gridNl – number of nodes in branch lkj – coefficient of coincidenceSnT – grid transformer rated power.

k=N k=N

k=N k=N

Ng Nl

Ng k=Nl Ng k=Nl


60

• in the case of application of a control system, which conditions the source output on the power flow inthe transformer:

SnG,1 ≤ SnT + kj × ∑l=1 (∑k=1 SMO,l,k) – ∑l=1 (∑k=2 SnG,l,k ) ≤ 2SnT (11)

Dependencies (9) and (10) refer to the Connect & Forget principle. At a given moment of time the source may output any power (but below, or equal to, rated output SnG,l) and the output does not have to be correlated with the power at the same time consumed by loads.

From the grid (grid operator) viewpoint, including loss of power (energy), the connection is “invisible.” It may cause some increase in the voltage levels, resulting from the reduced voltage drop on the grid transformer (HV/MV or MV/LV). This will depend on the source controller’s operating mode (control of voltage, reactive po-wer, or power factor) and the controlled quantity’s setpoint.

Dependence (11) refers to an arrangement with a control system of the source active power output (the Connect, Control Locally, and Forget principle). The source active power output controller/limiter should main-tain here the flow of the set power through the transformer to the MV grid, at a level below, or equal to, the gridtransformer’s rated power.

3. AN EXAMPLE OF ENERGY SOURCE CONNECTION TO RADIAL GRID

3.1. Grid modelThe following describes an example referring to an MV radial grid (or operated as such). The structure of

the subject grid is shown in Fig. 2. The subject the MV grid consists of 2 branches: 3xYHAKXS 150 mm2 cable lines with nodes spaced at 2 km. The following cable line specification is adopted: R’ = 0.268 Ω/km, X’ = 0.199 Ω/km, C’ = 0.27 µF/km, I = 355 A, I, , = 14.1 kA.

Four load nodes are situated on each radius. For the sake of simplicity, each node features the same diur-nal load variation presented in Fig. 3. The seasonal variation of the load characteristics is neglected here.

This grid is interconnected with a HV grid by a single high voltage transformer with the following rating: SnT = 16 MVA, 9= 115/15.75 kV/kV uk = 11%, ΔP cun = 91.5 kW.

It is assumed that there are four energy sources in the grid, i.e. generators G2, G3, G4 and G5. They are connected through set transformers to load nodes 8, 9, 10 and to the MV switchgear buses (node 6). The set transformer power ratings are adjusted to the source powers.

The operation of the HV/MV transformer’s voltage controller is neglected in this discussion. It is assumed, however, that the transformer’s HV voltage is constant and equal to 1.04, in relative units (RU).

k=Nl Ng k=Nl

Fig. 2. Example MV grid diagram

GGGG

Q

1

23 4 5

6

7 8 9 10

11 12 13 14T


61

Fig. 3. Diurnal variation of node loads, tgφ = 0.1

For an object so specified the grid operation is presented below in the form of the levels of the grid items’voltages and loads, whereby sources are connected to the grid according to the principles described above, i.e. Connect & Forget and Connect, Control Locally & Forget.

3.2. Grid with sources connected according to the Connect & Forget principleAccording to the relations presented in the previous section, the maximum outputs of sources connected

to a power grid according to the principle depend on the sequence of their connection to the subject grid ra-dius. This example assumes that the first source was attached to node 10. This has the effect that the source’srated (maximum) power output is greater than the rated (maximum) power outputs of the sources connected to nodes 8 and 9. These outputs, as defined by dependencies (1) for a source at the end of a branch and (6) forsources “inside” a branch, equal, respectively: P5 = 9.65 MW, P3 = 0.87 MW, P4 = 0.87 MW. The rated output of the source connected to the MV switchgear was calculated after (9). It equals P2 = 8.09 MW.

Two source operation variants, i.e. generator voltage and power factor control, are considered here. Figu-res 4 and 5 show the voltage variation and the loads of the grid’s items for the source operations under voltage control (adjustment in nodes 2, 3, 4 and 5, i.e. on the generator buses), where the setpoint voltage was equal to 1.01 RU (denoted in relative units).

Fig. 4 presents the voltage levels calculated for each hour of the day. The upper bundle of curves presents the diurnal variation of voltages in nodes of the branch to which energy sources are connected (the lower branch in Fig. 2), while the lower bunch of curves presents the voltages in nodes of the branch with no energy sources (the upper branch in Fig. 2).

A typical grid effect is evident here, i.e. voltages in branches without sources decrease as the distance from the supply node (e.g. node 6) increases, and voltages in branches with energy sources increase. It also shows a much smaller diurnal variation of voltages, as a result of changes in the load of the nodes in the source branch with sources, compared to branches without sources.

Fig. 5. shows the relative, i.e. related to the permanent current limit of the lines and transformer, load from grid items throughout a day. Here we can see that connection of sources in accordance with the Connect & Forget principle causes change in the direction of the power (energy) flow in the branch with sources. Po-wer flows toward the MV switchgear (positive values indicate a flow from node 6 to the loads, negative values – a flow toward node 6). Here we can see that in periods of the receiving nodes’ minimum load the flows inbranches 9-10, 8-9, 7-8, and in the HV / MV transformer (branch 1-6) are close to the items’ ratings. At other times the loads of grid items are below their ratings.

It is worth noting that the load of branch 6-7 is also below the rating, also at the minimum grid load. This follows from the fact that it is the difference between the power flowing in branch 7-8 (the limit for the branch)and the power consumed in node 7. This also means that in branch 6-7 there is a certain transmission capacity margin, and a power connection margin in node 7, i.e. this node might accommodate an energy source.

0,0

0,5

1,0

1,5

2,0

0 6 12 18 24

So

[MV

A]

T [h]


62

Fig. 4. Voltage distribution in grid nodes (node signatures: 1 - No. 1, 2 - 6, 3 - 7, 11, 4 - 8, 12, 5 - 9, 13, 6 - 10, 14, the upper curves cor-respond to nodes No: 7, 8, 9, 10, and the lower curves to nodes No: 11, 12, 13 and 14, the sources are under voltage control: U = 1.01)

0,940,960,98

11,021,041,06

1 2 3 4 5 6

U [j

.w.]

Węzły

Fig. 5. Grid item loads in the Connect & Forget variant (the sources under voltage control: U = 1.01)

As regards the operations of sources under power factor control, in the quantitative terms the situation changes somewhat, while quality-wise the grid operations is similar to that discussed above. This article assu-mes that energy sources operate at the power factor equal to tgφGz = -0.2, which means that they consume reactive power from the grid. This operating mode (contrary to requirements of system operators, who demand sources to operate at the reactive power equal to zero) reduces the voltage increase caused by inputting power to the grid, but increases the HV/MV transformer’s reactive power load and thus increases the voltage drop in the transformer.

As seen from comparison of the voltage variations in Fig. 6 and in Fig. 4, the source operation under power factor control leads to a substantial, albeit still permitted, increase in the voltage variation in load nodes (and at the recipients). It therefore can be concluded that the sources operation under voltage control, in terms of voltage variation is justified, although not absolutely required.

Comparison of Fig. 7 and Fig. 5 shows no major differences, including quantitative differences. In load valleys the power flows in the grid items situated upstream of sources are close to their ratings.

-1,5

-1

-0,5

0

0,5

1

0 6 12 18 24T [h]

I(1,6)

I(6,7)

I(6,11)

I(7,8)

I(8,9)

I(9,10)

Fig. 6. Voltage distribution in grid nodes (node signatures: 1 - No. 1, 2 - 6, 3 - 7, 11, 4 - 8, 12, 5 - 9, 13, 6 - 10, 14, the upper curves corre-spond to nodes No: 7, 8, 9, 10, and the lower curves to nodes No: 11, 12, 13 and 14, source under power factor control: tgφGz = -0.2)

0,920,940,960,98

11,021,041,06

1 2 3 4 5 6

U [j

.w.]

Węzły

Nodes

Nodes


63

Fig. 7. Grid item loads in the Connect & Forget variant (sources under power factor control: tgφGz = -0.2)

3.3. Grid with sources connected according to the Connect, Control Locally, and Forget principle

Connecting sources to distribution grids in accordance with the Join & Forget principle leads to an incre-ase in the use of their transmission capacities. However only at a grid’s minimum load the capacities are used fully, or rather to the greatest extent. Otherwise, i.e. as a matter of fact in most of a day, year, or operational period, these capacities are not used.

Connecting sources in accordance with the Join, Control Locally and Forget principle enables increase in the grid use. This is shown below.

In the case herein discussed adjusted are the currents in the grid sections situated upstream (as seen towards the HV/MV transformer) of the node, to which the given source is connected. This is achieved by adjusting the so-urce’s active power output to the grid. This means, for example, that the maximum permanent power permitted in branch 9-10 is maintained by controlling the active power output of the source in node 5 (generator G5), the power in branch 8-9 is maintained by controlling the active power output of the source in node 4 (G4), the power in branch 7-8 is maintained by controlling the active power output of the source in node 3 (G3). However, the power in the HV/MV transformer (branch 8-9) is maintained by controlling the active power output of the source in node 3.

In the first variant the sources operation under voltage control was considered. It is assumed here thatthe G2 generator voltage setting (node 2) is 1.08 RU. It is intended to limit the reactive power flow in the HV/MVtransformer. Lower voltage, equal to 1.01 RU, is maintained in the other sources.

As follows from Fig. 8, the voltage distribution in the grid is then similar to that shown in Fig. 4. One can see a higher voltage on the transformer’s MV buses (node 6) and slightly less variable voltage in the branch with energy sources.

The use of the grid transmission capacity increases, as shown in Fig. 9. The transmission capacities of the branches, downstream of which sources are situated, are fully utilised. Currents in branches 1-6, 7-8, 8-9 and 9-10 are equal to the ratings. The flow in branch 6-7 is below the setting, for the reason previously discussed.

-1,5

-1

-0,5

0

0,5

1

0 6 12 18 24T [h]

I(1,6)

I(6,7)

I(6,11)

I(7,8)

I(8,9)

I(9,10)

Fig. 8. Voltage distribution in grid nodes (node signatures: 1 - No. 1, 2 - 6, 3 - 7, 11, 4 - 8, 12, 5 - 9, 13, 6 - 10, 14, the upper curves correspond to nodes No: 7, 8, 9, 10, and the lower curves to nodes No. 11, 12, 13 and 14, the sources are under voltage control: Uz2 = 1.08, Uz3 = U z4 = U z5 , = 1.01)

0,970,980,99

11,011,021,031,041,051,06

1 2 3 4 5 6

U [j

.w.]

WęzłyNodes


64

Fig. 9. Grid item loads in the Connect Control Locally & Forget variant (sources under voltage control: Uz2 = 1.08, Uz3 = Uz4 = Uz5 = 1.01)

-1,5

-1

-0,5

0

0,5

1

0 6 12 18 24T [h]

I(1,6)

I(6,7)

I(6,11)

I(7,8)

I(8,9)

I(9,10)

Fig. 10. Active power outputs from sources to grid (sources under voltage control: Uz2 = 1.08, Uz3 = Uz4 = Uz5 = 1.01)

Fig. 10 shows the active power variation required to achieve the effect shown in Fig. 9. Based on the waveforms from this figure the power outputs of the sources required to obtain the presented condition maybe calculated. These powers should be equal to the maximum for the source read out from Fig. 9. In this exam-ple they are equal: P2 = 20 MW, P3 = 2 MW, P4 = 2 MW, P5 = 10 MW. These powers, listed here rounded up, correspond to those resulting from formula (3) for generator G5, from (7) for generators G3 and G4, and from (11) for generator G2.

Fig. 11 shows the reactive powers generated by sources, stemming from the preset source voltages. It can be seen here that generator G2 outputs, and generator G5 receives, a rather large reactive power. Reactive power generated by generator G2 is consumed to cover the reactive power absorbed in the upper branch of the grid and thereby reduces the reactive power flowing through the transformer HV/MV (Fig. 12 shows this).Therefore the reactive power output from this generator to the grid is justified, and even necessary. However,high consumption of the reactive power by generator G5 is not justified. This consumption may be reduced byincreasing the voltage setpoint. For example, increasing voltage Uz5 to 1.08, results in a decrease of the reactive power consumed to about 1 MVAr, but this will increase the voltage in node 10 to approx. 1.07 RU.

Reactive power flows in the HV/MV transformer may also be controlled by a capacitor bank connected tothe MV switchgear’s buses, or by the transformer (voltage ratio control). In either case, however, the adjustment will be discontinuous and relatively slow.

-6-4-202468

10

0 6 12 18 24

Q [M

var]

T [h]

Q2

Q3

Q4

Q5


65

Fig. 11. Reactive power outputs from sources to grid (sources under voltage control: Uz2 = 1.08, Uz3 = Uz4 = Uz5 = 1.01)

-6-4-202468

10

0 6 12 18 24

Q [M

var]

T [h]

Q2

Q3

Q4

Q5

Fig. 12. Active and reactive power flows through HV/MV transformer (sources under voltage control: Uz2 = 1.08, Uz3 = Uz4 = U z5 = 1.01)

Grid conditions related to source operation under power factor control are shown in Figures 13 ÷ 16. In this example it is assumes that sources operate at a set power factor equal to tgφ Cz = 0. This corresponds to the present requirements of system operators and to the system management practice.

Based on the waveform from Fig. 15, the minimum source power outputs theoretically required to obtain the proposed condition are equal to: P2 = 16 MW, P = 2 MW, P4 = 2 MW, P5 = 11 MW. Here the generator G2 maximum output is seen as much smaller compared to the previous example.

Due to the lack of reactive power output to the grid from generator G2 (tgφG = 0) the reactive power flowsthrough the HV/MV transformer are quite large (Fig. 16). This results in a voltage drop in the HV/MV transfor-mer larger than in Fig. 8 (operation under voltage control), and (as has already been mentioned here) less active power that may be output to the grid from generator G2 (Fig. 15 as compared to Fig. 10).

The generator G2 output, here below that obtained in the previous example, can be increased to the level shown in Fig. 10 by a system minimising the reactive power flow in the HV/MV transformer, e.g. by a controlled capacitor bankconnected to the MV switchgear. Use of the transformer for reactive power flow adjustment (theoretically possible) is notjustified here, because then the subject grid will be lacking an item responsible for maintaining voltage (control).

-5

0

5

10

15

20

0 6 12 18 24

P [M

W],

Q [M

var]

T [h]

PT

QT

0,940,960,98

11,021,041,061,08

1,1

1 2 3 4 5 6

U [j

.w.]

WęzłyNodes

Fig. 13. Voltage distribution in grid nodes (node signatures: 1 - No. 1, 2 - 6, 3 - 7, 11, 4 - 8, 12, 5 - 9, 13, 6 - 10, 14, the upper curves correspond to nodes No: 7, 8, 9, 10, and the lower curves to nodes No. 11, 12, 13 and 14, source operate under power factor control: tgφ Gz= 0)


66

The reactive power flow through the HV/MV transformer may also be reduced by an increase in the gene-rator G2 power factor or reactive power output setpoint. The feasibility of this action is however limited by the voltage increase at the end of the branch with power sources.

Fig. 14. Grid item loads in the Connect, Control Locally & Forget variant (sources under power factor control:

-1,5

-1

-0,5

0

0,5

1

0 6 12 18 24T [h]

I(1,6)

I(6,7)

I(6,11)

I(7,8)

I(8,9)

I(9,10)

Fig. 15. Active power outputs from sources to grid (sources under power factor control: tgφ Gz= 0)

0

5

10

15

20

0 6 12 18 24

P [M

W]

T [h]

P2

P3

P4

P5

Fig. 16. Active and reactive power flows through HV/MV transformer (sources under power factor control: tgφ Gz= 0)

4. ENERGY LOSSES

The issue of active power and energy loss in a grid is relevant to the above considerations and very impor-tant for distribution grid operators.

The grid capacity use maximisation up to the level resulting from the Connect, Control Locally & Forget or Connect & Control Actively principle obviously leads to an increase in the active power and energy losses

-10-505

101520

0 6 12 18 24

P [M

W],

Q [M

var]

T [h]

PT

QT


67

in the grid. This increase is expressed in power or energy units and results from the fact that power losses are a product of the square of the current in the respective power line sections. For example, losses in a power line at the level of utilization of 50%, after expanding its use to 100%, will increase fourfold.

Connection of distributed generation sources to a distribution grid does not necessarily lead to increased power and energy losses, since connecting sources of a certain output (less than specified in Chapter 2) leads in fact (can lead) to reduced grid load and hence to reduced power and energy losses.

For a grid of any structure the issue of grid energy loss minimisation may be formulated as follows. The current (power) outputs from sources to a system are looked for, for which the electricity (power) loss is lower than for the same grid without these sources. This issue is formalised in the following formula:

EbezG = ∑(∫T1 3 Ii,j Ri,j dt) > ∑ (∫T1

3 Ii,j – IG,j )2 Ri,j dt)= EzG (12)

where:Ii,j – current in branch i-jIG,j – current output from generator to node jRij – resistance of line i-jt,T1,T2 – time

If a branch of the radial grid as in Fig. 1 is considered subject to the following simplifying assumptions:• Variations of powers received in each load node are the same and have the form as shown in Fig. 17.

Current drawn by the load takes two values: maximum IM and minimum Im. Duration of current IM equ-als TM, and duration of current Im equals Tm.

• Power output to each node is constant in time and the same at each node of branch IG,j = IG.• Resistance Ri, j (and reactance) of each branch is identical, then formula (12) takes the form:

∑n=1 (nIM)2 TM + ∑n=1 (nIm)2 Tm > ∑n=1 (nIM – nIG)2 TM + ∑n=1 (nIm–n IG)2 Tm (13)

T2 T22

N–1 N–1 N–1 N–1

Fig. 17. Assumed example variation of distribution node loads

Inequality (13) is satisfied for a source current in the range:

(14)

Function f(IG) is equal to the difference between the left and right sides of formula (13) and takes the minimum for the sources current IG equal to:

(15)

This current may be converted to load (e.g. for comparison with formulas from Chapter 2) by multiplying it by the rated voltage and 1.73.

TMTm

IM

Im

I

t0

IG

T


68

The change in energy loss, i.e. the ratio of the right to left side of dependence (13), resulting from the operation of the sources that generate current IGopt of the value as in formula (15), is equal to:

(16)

Fig. 18 shows the dependence of the optimal source current IGopt and the ratio of energy losses in the system with and without sources, where the sources’ current output to the grid IGopt, is defined by formula (15). It is seen here that when the minimum load current Im is equal to zero, the so-called optimal source current IGopt should be equal to half the maximum current IM. Then the energy losses are reduced twofold, i.e. to 50% loss in the system without sources. As the minimum current Im increases, also the so-called optimal source current increases and the energy losses in the grid (subject branch) decrease.

In the extreme case, i.e. when Im = IM, the optimal source current becomes equal to current, IG = IM. then the energy losses in the grid fall to zero. This is because the source currents become equal to currents received in load nodes.

Fig. 18. Dependence of optimum source current IGopt and relative change in load EzG/EbezG on the ratio of currents I m / IM for TM = Tm

On the other hand, if an energy source is operated in a system only periodically, then the above dependen-cy is modified depending on the periodic variation of the source’s operation.

Let’s assume that the energy sources in the subject grid (branch) are wind turbines, which operate at times of increased energy demand, i.e. during the day. This is again a certain simplification. Yet it is reasonable, since during the night the wind speed decreases and so in fact does the wind power generation. Let’s assume also that:

• wind turbines operate at a constant power output (constant current IG)• wind power plan operation duration is TG and it does not exceed TM.

Then dependency (12) takes the form:

∑n=1 nIM TM > ∑n=1 (nIM – nIG)2 TG + ∑n=1 (nIM)2 (TM – TG) (17)

There is no minimum current Im, in this formula, because the energy loss caused occurs in the grid with no energy sources as well as with the subject sources.

Function f(IG) is equal to the difference between the left and right sides of formula (17) and takes the minimum for the sources current IG equal to:

N–1 2 N–1 N–1

EzG/EbezG

IGopt


69

IGopt = IM (18)

Equation (18) is intuitively legitimate, since the reduction of energy losses in the grid (branch) in the subject condition will take place only in the periods of the sources’ operation. The maximum reduction in energy loss occurs when the sources supply their “own” loads, i.e. the energy flow in the grid branches is minimized.

Change in the energy loss, i.e. the ratio of the right to left side of formulas (18, as well as 12), resulting from the operation of sources that generate current IGopt, as in equation (18), will then be:

(19)

Fig. 19 presents the ratio of energy losses in a system with and without sources, where the sources output to the grid the so-called optima current defined by equation (18), when the time of the sources’ operation is equal to the grid’s peak load: TG = TM = Tm. It is seen here that with the increase in the minimum current Im the energy losses also increase. This effect is opposite to that shown in Fig. 18.

(19)

Fig. 19. Dependence of relative change of energy loss EZG /E on the ratio of currents Im/IM for TG = TM = Tm

Summarising the above it should be noted that the connection of distributed generation can lead to a substantial reduction in energy losses in the distribution grid.

In the case of sources operated periodically during 24 hours, and specifically during the daytime, such as, for example, wind power plants (apart from the storm front passing periods) or photovoltaic sources, the maximum reduction in energy loss occurs when the sources operate with an output close to the maximum lo-ad power2..

However, as regards sources that operate (may operate) during 24 hours with a constant power output, such as biogas, biomass, and liquid fuel fired plants, the maximum reduction in energy loss occurs when the sources operate with outputs close to the weighted average of the maximum and minimum loads. For simplicity, the arithmetic average of the minimum and maximum loads may be adopted here.

5. SUMMARY

Growth in electricity demand in excess of the capacity of the grid transformer in a grid can be realized by replacement or addition of an HV/MV (or MV/LV) grid transformer, or by installation of energy sources in the grid.

2 The above statements are true subject to the adopted assumptions. Because of the actual load variation, as well as of the output variation of turbu-lent source, these have to be treated as a certain approximation.

EzG/EbezG


70

Energy sources may be installed in a grid up to the level derived from the dependence presented in Chapter 2. The grid transformer replacement is then not required. The dependences determining the maximum power output of a source that may be connected to a grid node refer to the condition of absence of the distri-bution system operator’s active influence on the grid (including the sources). This condition is convenient forthe system operator because it does not engage the operator in active influence on the system, either throughdirect control of sources, or through system services.

Where source(s) is (are) connected only deep in a grid, i.e. in nodes 2 to N, demand for energy in the grid may grow without having to replace the grid transformer with a larger unit or to build a new node to power the grid up to the total maximum load of the existing branches 1-2, i.e. of the first sections of the existing grid. Thisgrowth can not exceed the aggregate output power of the sources connected to the grid.

However, after connecting a source(s) to the switchgear buses on the side of the grid transformer’s lower voltage, the output power of the electricity source(s), provided that the existing grid infrastructure is not modi-fied, may be increased to a level limited by:

• permanent and short-circuit current-carrying capacity of the switchgear and control gear assembly• short-circuit strength of the existing switchgear and transmission lines (cables or overhead wires).It is worth noting that the extension of a grid (line) as a result of increased demand may cause an increase

in the power output of the sources which may be installed in the grid without having to implement a grid capex project, and are connected according to the Connect & Forget or Connect, Control Locally & Forget principle).

Any further growth in demand in the grid requires extension of the supply node or further increase in the output power of the sources connected to the grid. The increase in power sources, while maintaining the existing HV/MV (or MV/LV) grid transformer, essentially realizes the idea of grid capital expenditure replace-ment with a system service. This is then a source connection of the Connect & Control Actively type, where this control may be provided not only on-line, but it must assume the system service form3. Why? Because the most important requirement to be met by an electricity source is its capability of uninterrupted (reliable) operation in defined periods of time. It is required here, i.e. when the maximum power consumed by the loads exceeds therated power of the grid supplying transformer (apart from the possibility of the transformer’s periodic overlo-ading), that the source operation is more reliable than that of large generating units in the transmission system. This is due to the lack of redundancy for a source so situated. However such redundancy may be available, and in a number of ways. The basic way seems the operation of a sufficient number of generating units in the grid assources subject to the operator’s control.

Energy supply reliability should be stipulated by the agreement between a prosumer (energy source owner which is also an energy recipient, i.e. a producer cum consumer) and the grid operator.

Energy supply reliability may also be increased by employing energy storage technologies. These may be deployed at the prosumer’s, producer’s, or consumer’s site. From the viewpoint of grid operation and the de-pendences presented in Chapter 2, the energy storage location is irrelevant. It may, however, be relevant from the control quality and control system cost viewpoint. The energy that can be collected in an energy storage adjusts the minimum power that may be drawn in a grid node, i.e. the value of SmO,i. The application of an energy storage to increase the minimum power drawn in a node increases the power output of the sources that may be connected to the grid.

The burden of grid control may be entirely or partially transferred to an energy source, i.e. to a prosumer. This control can be provided only by the appropriate definition of the characteristics of selected control systemsof the prosumer’s generation unit, or by defining these characteristics and the grid operator’s concurrent on--line influence on the grid’s selected items. In these regulatory processes grid technologies of the smart grid type may be, and will be used.

An element, which in the future may materially affect the operation of this type of grid, will be the DSM (demand side management) system whereby the response to an emergency grid condition should focus on reducing the power consumed, while in normal operating conditions on “flattening” the load curve. Measuresaimed at the load curve flattening may be either active, i.e. direct control of the loads at an energy consumer’ssite (e.g. turn on in an off-peak period), or passive, such as influencing a customer by way of tariffs.

3 No services of this type are now provided in distribution networks because of legislative considerations.


71

REFERENCES

Summarizing the above, it may be concluded that the question arising from the title of the article: whether to operate a power grid up to its maximum, or to an extent that minimises the energy losses – still remains open. It may be answered by an economic analysis that takes into account the costs of grid extension, including the avoided costs.

It is certain, however, that in the existing power grids the connection of distributed generation sources in appropriate locations and with adequate rated powers will lead to a reduction of the power and electricity losses. Thus, it will be able to bring some financial benefits for distribution companies. It may also be concluded that the stage of distribution grids’ development involving the maximisation of their use (as a result of their saturation with distributed generation), will be preceded by the stage of minimizing the energy losses (as a result of limited saturation with distributed generation).

1. Kulczycki J., Rudziński M., Straty energii jako nieodzowne potrzeby własne sieci [Energy losses as grid’s indispensa-ble needs], Acta Energetica01/2009.

2. Tomczykowski J., Frąckowiak R., Gałan T., Przebiegi obciążeń odbiorców typu gospodarstwa domowe [Load curves of household-type electricity consumers], Conference Aktualne Problemy w Elektroenergetyce [Present Issues in Power Engineering] APE ’09, Jurata 3–5 June 2009.

3. Kot A., Kulczycki J., Szpyra W.K., Możliwość redukcji strat w sieciach dystrybucyjnych średniego napięcia poprzez optymalną lokalizację rozcięć [Feasibility of the losses reduction in medium voltage distribution grids by the optimal loca-tion of openings], Acta Energetica 02/2009


72


Jacek MareckiGdańsk / Poland

Obtained his bachelor (inżynier) degree in 1952 at the Faculty of Electrical Engineering, Gdańsk University of Technology. Worked as a designer of industrial power and CHP plants (1952–1954) and after obtaining MSc (magister inżynier) degree in 1954 went to work at the construction site of Czechnica power station in Wrocław. As a Ford Foundation scholarship holder attended one-year postgraduate study programme in nuclear power engineering at the Royal College of Science and Technology in Glasgow (1958–1959) and then went on to a scientific traineeship period at Electricité de France in Paris (1962). Obtained his doctoral degree (1961) and later DSc (doctor habilitatus, 1966) at the Faculty of Electrical Engineering, Gdańsk University of Technology. In 1971 awarded title of Associate Professor and in 1979 – Full Professor. In 1991 appointed corresponding member of the Polish Academy of Sciences, since 2004 full member. In 2010 appointed member of the Academy of Engineering in Poland.As a long-standing academic teacher at the Gdańsk University of Technology (1959–2005) has significant achievements in educating power engineering experts, particularly specialists in combined heat and power generation, nuclear power, and complex power system management. Supervised fourteen doctoral theses. Author or co-author of more than 200 publications, including ten monographs, scientific studies, as well as four academic textbooks. Monograph “Combined Heat and Power Generating Systems” published in the UK has been frequently quoted. Academic textbook “Podstawy przemian energetycznych” on principles of energy conversion processes has already had three editions and is in use at faculties of electrical engineering of technical universities all over Poland.

Jacek Marecki / Gdańsk University of Technology

73

MODERN NUCLEAR POWER TECHNOLOGIES1


1. INDTRODUCTION

Modern power generation technologies are among the most innovative areas of science and technology. Currently we can observe a renaissance of one of the power engineering branches – nuclear power technolo-gies. New power units of this class are now planned to be constructed in many countries of the world, including Poland.

Numerous studies carried out in recent years and presented at various Polish and international confe-rences, like those organized by the “Poland 2000 Plus” Forecast Committee affiliated with the Presidium of the Polish Academy of Sciences and the Committee on Energy of the Polish Academy of Sciences, have highlighted the necessity to construct nuclear power stations in Poland. The authors argue that starting a nuclear power programme in our country is required not only from an energy management point of view, but is also necessary for economic and environmental reasons.

In order to assure power supply security in a country it is necessary to construct new base-load power plants to meet the expected increase in electricity demand, and do so in a way which is feasible from the following points of view:

• Energy – meaning taking advantage of available assets and diversification of primary energy sources• Economy – minimizing discounted energy generation costs• Environment – meaning compliance with European Union legal framework and taking into account

forecasted CO2 emission allowance prices.Development of power generation technologies is of course related to evolution of the main energy

conversion equipment types – which in the case of nuclear power stations means reactors. Therefore this paper presents:

• Features of main types of modern nuclear power reactors, focusing on used nuclear fuel, moderator and coolant

• Current count of nuclear reactors installed in operational nuclear power stations all over the world, showing total electrical output installed in power units with individual reactor types

• Results of recent economic studies carried out for base-load power plants scheduled to be commis-sioned around the year 2030: nuclear with pressurized water reactors, thermal fired with coal or lignite and combined-cycle units fired with natural gas.

2. FORECASTS FOR NUCLEAR POWER DEVELOPMENT IN 21ST CENTURYIt is estimated that by the year 2050 the world’s population will have increased to approximately 10 bil-

lion, with almost the entire growth concentrated in developing countries. Therefore the global energy demand

Modern Nuclear Power Technologies

1 This article is a reworked variant of the paper presented at the plenary session of the Science and Technology Conference – Modern Power Engi-neering Technologies and Machines held in Kraków on 15-17 September 2010.

Abstract

The paper discusses possibilities of using modern nuclear power technologies in power stations. Forecasts for nuclear power development until 2050 have been pre-sented. Shares of individual reactor types in total power installed in operational nuclear power plants have been given. It is forecasted that the first Polish nuclear power

plant will utilize one of three reactor types: pressurized water reactors (PWR), boiling water reactors (BWR) or heavy water CANDU reactors. Results of comparative analyses on electricity generation costs in nuclear and thermal power plants (running on hard coal, lignite or natural gas) have been presented

74

might double in reference to the year 2010. There is no doubt that the supply of energy in general and electricity in particular is among the primary human needs, not only in industrialized countries, but also – with increasing intensity – in developing countries.

Figures 1 and 2 present results of global electricity consumption forecast until 2050 carried out by the International Institute for Applied Systems Analysis (IIASA) and presented at one of the World Energy Council’s congresses [1]. It is predicted that the global electricity consumption in 2050 will reach approximately 40 PWh (40 billion MWh) – which is 3.5 times more than in 1990.

Most probably the nuclear power sector will cover a significant portion of that increase. Recent publica-tions, the outcome of which has been presented in various studies such as [2-5], highlight the following features of nuclear power technologies:

• Availability of large deposits of uranium and – when needed – thorium which have no other applications outside the power industry

• High security of supplies as creation of long-term (many years) strategic fuel reserves is easy• Practically no emissions of carbon dioxide (CO2) and other harmful substances• Stability of electricity generation costs, resulting from the fact that nuclear fuel cost is only some 5% of

the total electricity generation cost in a nuclear power station.

OECD

Transf.

Developing

103 TWh

11.8

21.5

41.0

18

7.2

11.9

17.22.9

5.8

2, 46.72.2

1990 2020 2050

Fig, 1. Global electricity consumption forecast according to [1].

Fig. 2. Global electricity consumption forecast (%) according to [1].

1990 2020 2050

OECD

Transf.

Developing

61

20

19

55

31

14

44

42

14

100%

North America

EU + Switzerland

Commonwealth of Independent States

Asia

South America

Africa

Fig. 3. Nuclear power industry around the world in late 2008 – installed capacity of nuclear power units [6]

Total capacity installed in nuclear power stations 371.8 GW (438 units)

35.2

82.8134.8

114.5 2.7

1.8


75

Figures 3 and 4 present the status of the nuclear power industry around the world at the end of 2008 according to statistical data published by the International Atomic Energy Agency in 2009 [6]. The total electric capacity installed in 438 operational nuclear power generation units was 371.8 GW, of which 134.8 GW was located in the European Union member states and Switzerland. The share of nuclear power plants in global elec-tricity generation in 2008 was 17.7%, while in the European countries mentioned above – around 37%.

The IIASA forecast mentioned above [1] states that – according to the scenario assuming strict environ-mental regulations and application of sustainable development principles – nuclear power generation capacity will increase to approximately 1120 GW in 2050 – thus tripling the current value. Those predictions are based on the assumption that development of nuclear power industry will focus mainly in the countries which are already operating nuclear power stations at the present day or are already planning to build such units.

This group of countries will be soon joined by Poland, which – after a period of lively discussion about the necessity to develop nuclear power, presented for example in a presentation of Polish experts at the World Ener-gy Council congress in 2007 [4] – is now proceeding to the execution phase of the government programme. Therefore there is a real possibility that by 2030 at least two nuclear power plants with an output of 1500-1600 MW each will be commissioned in Poland.

3. TYPES OF NUCLEAR POWER TECHNOLOGIES

Soon the issue of selecting power generation technology – meaning the type of reactor and auxiliaries – for the first Polish nuclear power plant will become a subject of studies, discussions and a public procurement procedure. For this reason the most important types of nuclear power reactors, described for example in the study [7], are briefy presented below. Table 1 shows commonly used English acronyms of the reactor types, which usually indicate types of moderator and coolant.

North America

EU + Switzerland

Commonwealth of Independent States

Asia

South America

Africa

Nuclear power generation: 2603 TWh (17.7% of total global generation)

239

504907

920

20 13

Fig. 4. Nuclear power industry around the world in late 2008 – electricity generation in nuclear power units [6]


76

Table 1. Nuclear reactor types according to [7].

Reactor type Acronyms Fuel Moderator Coolant

Light water pressurized PWR(APWR)(VVER)

Enriched UO2 Light water Pressurized light water

Light water boiling BWR(ABWR)

Enriched UO2 Light water Boiling light water

Heavy water PHWR Natural or enriched UO2 Heavy water Pressurized heavy water

Water-graphite LWGR(RBMK)

Enriched UO2 Graphite Light water

Gas-graphite GCR(AGR)(HTGR)

Natural uranium orenriched UO2

Graphite CO2 or helium

Fast breeder FBR Enriched UO2 + PuO2 - Liquid sodium

Fig. 5 shows shares of individual reactor technologies in the total installed capacity of operational nuclear power plants according to IAEA data of 2009 [6]. It turns out that the dominating technology is pressurized water reactors (PWRs) – with total electric capacity of 243 GW installed at 264 power units.

The second most popular technology is boiling water reactors (BWRs); units utilizing them have a total capacity of 85.3 GW – approximately 23%. The total capacity installed in nuclear power units utilizing both types of light water reactors is therefore 88% of the total value. Those reactor types utilize enriched uranium fuel in the form of uranium dioxide (UO2), and are cooled and moderated with light water (H2O).

The third most popular type – in terms of installed capacity and its share in the total capacity of all nuclear power plants in the world – is pressurized heavy water reactors – PHWRs. The total capacity of units with such reac-tors is more than 22 GW, 6% of the world’s total. Fuel for such reactors can be natural uranium, as heavy water (D2O) displays better nuclear properties, though they can use lightly enriched uranium in UO2 form as well.

It seems that selection of the reactor type for the first Polish nuclear power plant – and possibly also for the subsequent plants – will need detailed analysis of all technical and economic aspects of all those three tech-nologies. Possible reactor types can be:

PWR

BWR

PHWR

LWGR

GCR

FBR

Totally 438 units, 371.8 GW

PWR: 264243 GW65. 4%

BWR: 9485,3 GW

22.9% PHWR: 4422.4 GW

6.0%LWGR: 1611.4 GW

3.1%GCR: 189.0 GW

2. 4%FBR: 20.7 GW

0.2% Fig. 5. Types of reactors in operational nuclear power units in the world, late 2008 [6]


77

• Light water pressurized reactors PWR (APWR – Advanced Pressurized Water Reactors) offered by We-stinghouse or a French-German consortium (EPR – Evolutionary Power Reactor)

• Light water boiling reactors BWR (ABWR – Advanced Boiling Water Reactors) offered by General Elec-tric

• Heavy water reactors PHWR offered by Atomic Energy of Canada Ltd under the brand CANDU (Canada Deuterium Uranium Reactor).

Technical and financial analyses will be of key importance here. They should take into account both theinvestment cost of power plant construction, and all types of annual operation costs – fixed and variable, internaland external. In the case of fossil fuel power plants compared to nuclear units, external costs are mainly those generated by purchasing CO2 emission allowances.

4. ECONOMIC ASPECTS OF COMMERCIAL POWER PLANT DEVELOPMENT

One of the newest comparative analyses of electricity generation in various power stations was presented by M. Duda in his study [8] based on calculations carried out by Agencja Rynku Energii in 2009 and commissio-ned by the Ministry of Economy. This study investigated more than ten types of power stations scheduled for commissioning around the year 2020 and later – up to 2050, among them particularly important large base-load power stations:

• Nuclear power stations with III generation PWR reactors fuelled with enriched uranium• Condensing power stations fired with hard coal combusted in pulverized bed boilers equipped with flue

gas desulphurization and denitrification systems, and optional additional carbon capture and storage systems• Condensing power stations fired with lignite combusted in pulverized bed boilers with additional sys-

tems as above• Gas turbine combined cycle units fuelled with natural gas, with optional CCS systems.Table 2 and Figure 6 present the key results of calculations quoted in the study [8], carried out in Euro of

2005 (€’05). They clearly show the competitiveness of nuclear technology when compared to fossil fuel power plants, resulting from CO2 allowance costs.

It would be useful to extend this type of analysis to other types of nuclear technologies, which could be utilized in the first Polish nuclear power plants. Among them are:

• Units with boiling light water reactors (ABWR and similar) fuelled with enriched uranium• Units with pressurized heavy water reactors fuelled with natural uranium, according to CANDU

technology.Thorough technical and economic analyses for various types of nuclear technologies should soon result

in the final selection of the reactor type and vendor for the first Polish nuclear power station, which is scheduledfor commissioning around the year 2020.


78

BIBLIOGRA PHY

Table 2. Economic comparison of commercial power stations to be commissioned around 2030 according to [8].

Power plant technology

Investment cost Fuel energy cost Electricity generation cost

€’05/kW €’05/GJ €’05/MWh

Nuclear PWR 2900 0.8 64

Hard coal 1600 3.8 98

Hard coal + CCS 2400 3.8 90

Lignite 1700 2.2 92

Lignite + CCS 2500 2.2 78

GTCC 800 11.2 102

GTCC+CCS 1200 11.2 104

1. Global Energy Perspectives to 2050 and Beyond. 17th Congress of the World Energy Council, Special Session 2, Houston 1998.

2. Marecki J., Wójcik T., Perspektywy energetyki jądrowej w XXI wieku. Perspektywy awangardowych dziedzin nauki i technologii do roku 2010. Komitet Prognoz „Polska 2000 Plus” przy Prezydium PAN, Warsaw 1999.

3. Marecki J., Duda M., Aspekty energetyczne, ekonomiczne i ekologiczne rozwoju elektrowni jądrowych, Systems 1/2, 2006, Vol 11.

4. Marecki J., Duda M., Kerner A., Why Should Poland Go Nuclear? 20th Congress of the World Energy Council, Rome 2007.

5. Program on Technology Innovation: Integrated Generation Technology Options. Electric Power Research Institute, 2008.

6. Nuclear Power Plants in the World. International Atomic Energy Agency, Vienna 2009.7. Chwaszczewski S., Technologie jądrowe w XXI wieku. Polityka Energetyczna, 2/2, 2009, Vol 12.8. Duda M., Aspekty ekonomiczne rozwoju elektrowni jądrowych, Spektrum 3/4, 2010.

64

78

929098 102 104€’05/MWh

Fig.. 6. Electricity generation cost for base-load power stations to be commis-sioned around 2030 according to [8].


PWR NPP

Lignite + CCS

Hard Coal

+ CCS

Lignite Hard Coal

GTCC GTCC + CCS

80


Krzysztof ŻmijewskiGdańsk / Poland

Works as a Product Manager for Energa SA in implementation of modern technologies and products. Graduate of Śląskie Techniczne Zakłady Naukowe high school in Katowice. Obtained his magister inżynier (MSc) degree at the Warsaw University of Technology. Worked as a customer relations expert for key customers of Netia SA. Experienced in sales, sales network management, customer relations, strategy planning, project management, marketing, B2C, B2B, particularly in the creation and implementation of new products.

Krzysztof Żmijewski / ENERGA-OBRÓT SA

81

VAWT – A KEY FOR DISTRIBUTED GENERATION

Krzysztof Żmijewski / ENERGA SA

INTRODUCTION

Why is fast development of modern technologies not directly translated into immediate application po-ssibilities? With this question I would like to draw your attention to the issue of using small vertical axis wind turbines (VAWTs) in the development of distributed microgeneration capacity.

Power generated at power plants, CHP plants and renewable energy sources is 70% purchased in a free market. The remaining output is sold according to specific shares, where output from CHPs is 20.6%, renewable 8.7% and gas-based generation 2.9%1.

DISTRIBUTED GENERATION – VERTICAL AXIS WIND TURBINES

My area of focus includes renewable energy sources which enable power generation in a distributed structure. The concept of distributed generation as such is not precisely defined and we can find various defi-nitions based on different classification methods: power output, connection method or generation technology. The most popular definition uses studies of the International Council on Large Electric Systems and the Electric Power Research Institute. In Poland distributed power generation is defined as small (installed capacity up to 50-150 MW) sources or plants connected directly to the distribution grids or located within a consumer’s own grid (behind the measurement unit), often generating electricity from renewable or unconventional sources, or – just as frequently – in cogeneration with heat2.

Distributed power generation is a very broad issue, which is affected by the evolution of the entire global energy market. As the demand for electricity grows, we observe various processes which greatly affect the energy market. The market is influenced by legal regulations, new technology development, climate control policy – including CO2 emission reduction commitments, safety of supplies and shrinking deposits of fossil fuels. Those trends will significantly affect end energy users in the nearest future. Development of distributed power generation creates a chance to make better use of renewable natural resources and – very significantly – de-velopment of end consumers’ awareness about energy issues. A short list of events which drive changes in the Polish energy market is presented here.3

1 www.cire.pl, Centrum Informacji o Rynku Energii web portal.2 Józef Paska, Multimedialny katalog generacji rozproszonej.3 Tomasz Sikorski, Edward Ziaja, Generacja rozproszona na tle obecnej struktury energetyki krajowej, Energetyka, December 2008.

Abstract

The paper presents the issue of micro power generation with wind turbines. The author has presented the most important factors to be taken into account when investigating the feasibility of micro power plants, shows the main features and parameters of vertical axis wind turbines (VAWTs) and challenges which need to be tackled when realizing a project involving this technology.

VAWT – a Key for Distributed Generation

82

Tab. 1. Key events which influence the Polish energy market.

Major Political event Major events for Polish power industry

1989 Programme documents for the Polish power industry

1990 Premises for the energy policy until 2010

1997 The Energy Law Act

2000 Premises for the energy policy until 2020

Application for EU membership 2002 Progress evaluation and revision of premises of the energy policy until 2010

EU Accession Treaty 2003

Poland joins the European Community 2004

Energy plans of European Community 2005 Polish energy policy until 2025

Directives and regulations of the European Commission and Council

2006 The Energy Law Act amended

2007 Polish energy policy until 2030

Application of distributed power generation may be the first step towards a new power generation model.Instead of centralized generation combined with expensive long-distance transmission and distribution, we sho-uld maximally use opportunities of generating power where it is consumed. An additional argument in favour of popularizing renewable energy sources is their pro-environmental character.

We should start the process of analyzing a distributed microgeneration project by selecting appropriate tools. There are multiple solutions for power and heat generation available on the market. The most popular ones are solar collectors, heat pumps, wind turbines and photovoltaic cells.

The most popular technologies (except conventional) which may be applied for distributed generation are:

• Combined heat and power systems based on: – steam turbines – reciprocating engines, – gas turbines – microturbines• Wind turbines – drum type – carousel type – rotor type – multi-blade type – propeller type – tornado type• Photovoltaic cells• Fuel cells• Small hydroelectric plant.

The issue of local energy source is very broad, and all those technologies can be applied. I have selected wind turbines as an example solution which can be used for distributed microgeneration.

Human civilization has been using wind power for practical applications for quite a long time. In practical terms we can convert wind energy into electricity or heat. The useful form of energy generated in this way will be an additional input to the whole energy system. There is plenty of information available on the wind farms which are planned to be developed in Poland and other European countries or are already operating all over the world4. There is less data however on electricity generation solutions for individual users. The wind farms

4 Renewable Energy in Poland, Polska Agencja Informacji i Inwestycji Zagranicznych, Warsaw 2009, (Invest in Poland, APAX Consulting Group).

Krzysztof Żmijewski / ENERGA-OBRÓT SAKrzysztof Żmijewski / ENERGA-OBRÓT SA

83

which are built in Poland can be used as an example of using wind power for electricity generation5. They should also be guidelines and reference for possible investors. Wind farm construction is not a solution for a person who only needs some electricity for their own consumption. The correct technology for individual users is small wind turbines which allow generating electricity on a small scale, for own consumption of a household or a small business.

A wind turbine for local electricity microgeneration should be first of all easy to install on a building or mast. Small weight and dimensions are important characteristics which influence the ability to install it on most building structures. Another important parameter which affects the efficiency is the wind speed at which the turbine starts to generate electricity. Bearing in mind the high price of a turbine, it is worth paying attention to its manufacturing quality and lifetime. Reliable evaluation and selection of a proper turbine is still a rather diffi-cult task. Most models offered on the market are new designs which have not yet worked for the whole lifetime declared by the manufacturers. Leading suppliers claim that this is no less than 25 years. Turbines dedicated for installation in urbanized areas or in rural areas must be safe for their environment; they should also have limited impact on that environment. Getting social approval and understanding for the pro-environmental character of such a power generation technology is a very important issue for a project.

5 Energetyka wiatrowa w Polsce, report, November 2009, Polish Information and Foreign Investment Agency.


Power curve for a 3 kW turbine

Wind speed (m/s)

Pow

er [

W]

84

Industry’s response to those expectations are vertical axis wind turbines (VAWTs). One of the key parame-ters used by their suppliers is power output given in watts.

Turbine manufacturers specify a nominal capacity for a device. In practice, however, generated electricity volume will depend on the characteristics of the design and real wind conditions for the selected site. Turbines of various outputs, ranging from 300 W to 20 kW, are available on the market. The most popular products are rated between 1000 and 6000 W. For a better understanding we may say that a 750 W turbine at an average wind speed of 5.5 m/s will generate approximately 2.2 MWh of electricity per year. For a 3 kW and 6 kW turbines it will be 6.5 MWh and 12.5 MWh respectively.

Vertical axis wind turbines have several unique characteristics which make them a good technology for widespread use:

• Low start-up wind speed – they start to operate at low wind speeds of 2-3 m/s• Higher efficiency than horizontal axis wind turbines• Reliable operation at variable wind speed and direction• Ability to install directly on building structures as well as on low and tall masts• Quiet operation• Braking system assuring safe operation even at high wind speeds• Electronic control• Simple design and easy installation.All those features make VAWTs suitable for operation in urbanized areas and cities. In order to provide ca-

pacity for own electricity demand first of all it is required to determine the value of that demand – for electricityor heat – and select appropriate output of the generation equipment. It is also necessary to select a favourable installation location and investigate conditions there. The evaluation process should include:

• Evaluation of the space required for a wind turbine installation• Check for tall obstacles in the area, which might create a barrier and distort general wind conditions• Evaluation of wind potential based on observation or a digital wind atlas• Determining correct mast height or any other support system for the turbine.When selecting an installation site it is also necessary to take into consideration the distance to the ne-

arest interconnection point to the electric grid. Gathering all that information enables making a decision about installation of a local and independent electricity source. The investor should investigate an option of installing the turbine on a building in a way which does not require obtaining a building permit. In other cases complex conditions imposed by the Construction Law must be met and regulations of the municipal spatial development plan need to be followed. Installing a turbine on a mast or building which requires a building permit significantlyincreases project cost and extends the scope of the required analyses.

Krzysztof Żmijewski / ENERGA-OBRÓT SA

85

CONCLUSION

An unregulated wind turbine market in Poland combined with rapid introduction of new turbine models poses a challenge for new legal regulations. The current law does not clearly distinguish small turbines from large systems used at wind farms. As a result installation of such equipment for own needs must meet the same requirements as applied for large scale applications. This is one of the key barriers in the development of such solutions for use by individual customers.

There is no doubt that the science, technology and legal regulations should support business initiatives promoting distributed power generation. The most important step is to quickly remove existing barriers and establish support mechanisms for quick development of a market based on distributed energy generation. This process will be beneficial to the natural environment as well as the security of energy supplies to the end custo-mers.


86


Maciej M. SokołowskiWarsaw / Poland

Graduate of American Law Center (The University of Florida Levin College of Law), Faculty of Law and Administration, University of Warsaw. Senator of the University of Warsaw. Holder of a scholarship from the Minister of Science and Higher Education for studying performance. Cooperates with Polish and foreign law offices, in the past trainee at RWE Polska, the Energy Regulatory Office, Ministry of Economy and the National Atomic Energy Agency. Currently holds the position of Executive Director at the Public Advisory Board for National Emission Reduction Programme and is an an expert of the International Federation of Industrial Energy Consumer of Europe. Author of publications and expert studies on Polish and European law in areas of energy, nuclear industry and climate, as well as domestic and international energy policy.

Krzysztof Żmijewski / Public Advisory Board for National Emission Reduction ProgrammeMaciej M. Sokołowski / Public Advisory Board for National Emission Reduction Programme

Krzysztof ŻmijewskiWarsaw / Poland

Lectured at the Polish-Japanese Institute of Information Technology and Krajowa Szkoła Administracji Publicznej. President of the Board of Polskie Sieci Elektroenergetyczne SA (1998–2001) and Polish National Energy Conservation Agency (1993–1998). Undersecretary of State at the Ministry of Construction Industry in charge of energy efficiency issues in the municipal sector (1991–1993). Representative of Poland at the committee of SAVE, the first EU programme involving Poland (1997– 2000). Former member of the National Development Board, advisory body of the President of the Republic of Poland. Currently Secretary General of the Public Advisory Board for the National Emission Reduction Programme which advises the Vice Prime Minister and the Minister of Economy, also independent consultant. Theoreti-cally and practically investigates issues of energy efficiency, energy policy, renewable energy sources and commercial power engineering. Works at the Department of Building Construction, Warsaw University of Technology.

87Power Grid Development in Poland in the Context of EU Climate

and Energy Package

POWER GRID DEVELOPMENT IN POLAND IN THE CONTEXT OF EU CLIMATE AND ENERGY PACKAGE

Krzysztof Żmijewski / Public Advisory Board for National Emission Reduction Programme Maciej M. Sokołowski / Public Advisory Board for National Emission Reduction Programme

1. INTRODUCTION

Power grid is an irreplaceable part of any power system. It interconnects all the elements and enables a system’s primary duty – providing electricity to the national economy. At the beginning of the 21st century nothing indicates that the share of electricity in energy balances could drop, so also the need for grid opera-tion is not going to diminish. This of course does not rule out development of new energy sources1, or research aimed at improving energy efficiency2, storage and transfer3.

The industrial revolution which started at the turn of the 19th and 20th centuries initiated a rapid populari-zation of electric equipment, requiring constant development of power supply systems. Those processes were interconnected with growing demands of societies and economies. In the 21st century further extension of po-wer grids is still a must – although now the extension has somewhat changed its meaning. The key issues are internal and external aspects of grid development – the ideas will be discussed further in this paper.

Another issue discussed by the authors of this study are regulations of the Climate and Energy Package, as well as Polish actions and the situation resulting from that package. Its impact on the investment needs in the Polish grid will be identified, together with development problems, obstacles and concepts for intensification of real actions in this field.

2. DUALISM IN GRID DEVELOPMENT

The internal aspect of a power grid development, defined as an internal (domestic) investment need, primarily consists of actions aimed at proper grid structure management by increasing its density and topology (physical development of transmission and distribution connections), improving existing grid management (in-stallation of modern grid control systems), and upgrading systems for operation monitoring and failure preven-tion (sensors and instruments installed at individual lines and stations). Internal grid development also includes maintenance operations and replacement of depreciated and obsolete power lines.

The internal development aspect results mainly from the national electricity demand, investment needs of individual investors and necessity to assure security of electricity supply to every consumer. This development should be carried out according to the plans created by each Distribution System Operator (DSO) and the Trans-mission System Operator (TSO) according to the regulations of the Act of 10 April 1997 – Energy Law4 (further referred to as the “Energy Law”). In Article 9c, Section 3 of the Energy Law, which defines the general duties

1 E.g. large-scale nuclear fusion power generation.2 For example by increasing efficiency of individual devices (at both generating and consumption side).3 E.g. research on high-efficiency, high-capacity batteries.4 Consolidated text of 2006, Dz.U. Nr 89, poz. 625 as later amended.

Abstract

There are many internal and external aspects affecting power grid development in Poland in the context of the Climate and Energy Package. Internal grid develop-ment aspects are mainly caused by the domestic electri-city demand, investments and the requirement to assure

security of supplies. External aspects of grid development result from international multilateral agreements and decisions made by international organizations. In this pa-per its authors try to explain those issues and share their evaluation of the general grid situation in Poland.

88

5 See M. Czarnecka, T. Ogłódek, Prawo energetyczne. Komentarz, Warsaw 2009, p. 234.6 Directive 2009/29/EC of the European Parliament and of the Council of 23 April 2009 amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading scheme of the Community. Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. Directive 2009/31/EC of the European Parliament and of the Council of 23 April 2009 on the geological storage of carbon dioxide and amending Council Directive 85/337/EEC, European Parliament and Council Directives 2000/60/EC, 2001/80/EC, 2004/35/EC, 2006/12/EC, 2008/1/EC and Regulation (EC) No. 1013/2006Regulation (EC) No. 443/2009 of the European Parliament and of the Council of 23 April 2009 setting emission performance standards for new passen-ger cars as part of the Community’s integrated approach to reduce CO2 emissions from light-duty vehicles.Directive 2009/30/EC of the European Parliament and of the Council of 23 April 2009 amending Directive 98/70/EC as regards the specification ofpetrol, diesel and gas-oil and introducing a mechanism to monitor and reduce greenhouse gas emissions and amending Council Directive 1999/32/EC as regards the specification of fuel used by inland waterway vessels and repealing Directive 93/12/EEC.Decision No. 406/2009/EC of the European Parliament and of the Council of 23 April 2009 on the effort of Member States to reduce their greenhouse gas emissions to meet the Community’s greenhouse gas emission reduction commitments up to 2020.OJ L 140, 5.6.2009.

of the operators5, Item 11 stipulates that a DSO is responsible for planning distribution grid extensions, while applying rules assuring equal rights of all system users and environment protection principles, and taking into account energy efficiency projects, electricity demand management and development of generation capacityconnected to a distribution grid. As for the TSO (Article 9c, Section 2, Item 4), the legislator makes it responsible for assuring the long-term ability of the power grid to meet justified needs of electricity transmission in domesticand international trade, including transmission grid development and – wherever applicable – construction of cross-border connections to other power systems.

There are also other significant regulations concerning operation of TSO and DSOs which directly andindirectly affect power grid development. One of these is Article 9c, Sections 2 and 3 on security of electricity supplies by assuring reliable operation of the power grid and appropriate transmission capacity in the trans-mission and distribution power grids, operation and maintenance of the grid and related equipment including interconnections with other grids, in a way which guarantees operational reliability, cooperation with other grid operators and power industry companies aimed at assuring cohesive operation of grids and coordination in their development, as well as reliable and efficient operation of those systems.

Those regulations are interconnected with stipulations of Article 16, Section 1, which state that power industry companies which operate in areas of electricity distribution or transmission shall create development plans for their business area, with focus on meeting current and future electricity demand, taking into account local spatial development plans and municipal development directions defined by a study of conditions anddirections of a municipality’s spatial development. According to Article 16, Section 3 a key part of those plans shall be projects of modernization, extension or construction of the grid and possible new electricity sources, including renewables.

It needs to be said here that in real life the requirements presented above are only enforced in a formal respect (e.g. development plan creation), while the material requirements – real modernization or investment actions – are not enforced at all.

In other words the Energy Law only imposes obligations to create development plans, but does not requ-ire them to be executed. There are no sanctions against inactivity in this respect (except for the general regula-tion which commits the Operator to accomplish its duties – Article 56, Section 1, Item 24).

The presented legal environment creates a framework for internal reasons for power grid development. By contrast, the external aspects result from circumstances which are not directly related to the domestic energy situation. The main drivers here are international multilateral agreements and decisions made by international organizations. The authors of this study would like to particularly highlight conditions specific for Europe anddecisions made by European Union entities, including regulations which affect the development and situation of the Polish grid and its extension. One such element is certainly the so-called Climate and Energy Package.

3. CLIMATE AND ENERGY PACKAGE

The Climate and Energy Package is a set of official documents6 enacted in 2009 with which the European Union imposed pro-climate and pro-environmental commitments in the form of compulsory quantitative targets


89

set for the year 2020. They are supposed to be reached jointly by all EU member states. Those targets are: gre-enhouse gas emission reduction, increase of renewable energy sources (further referred to as “RES”) share in the final energy balance, and decrease of energy consumption (the so-called 3 × 20% targets).

Starting in 2013, all efforts of the European Union aimed at greenhouse gas emission reduction in refe-rence to the 1990 level by 20% by 2020 will be divided between the sectors included in the EU ETS7 and those which are not parts of the system (non-ETS8):

a) 21% emission reduction in EU ETS sectors in reference to the 2005 level,b) approximately 10% emission reduction in non-ETS sectors in reference to the 2005 level.When combined, those actions should result in a total reduction of approximately 20% in reference to

the year 1990 and 14% to 2005. The European Commission expects larger reductions in the EU ETS sectors, as it considers appropriate measures to be more cost-effective than in non-ETS sectors9. Further commitments imposed by the European Union on its member states and related to the Climate and Energy Package are: de-creasing energy consumption by 20% compared to the EU forecasts for 202010 achieved by improving energy efficiency, and increasing the RES share to 20% of total EU energy consumption, including growth of the RESshare in the transport sector to 10%11.

It needs to be emphasized that only some of those regulations are compulsory.

Regulation Field Compulsory Scope

Directive 2009/28/EC RES YES European

Directive 2009/29/EC ETS YES National

Directive 2009/30/EC FUELS YES National

Directive 2009/31/EC CCS YES National

Regulation 443/2009/EC FUELS YES National

Decision 2009/406/EC non-ETS YES National

Directive 2006/32/EC EFFICIENCY NO -

A significant issue for Poland which results from the stipulation of the Climate and Energy Package is so-cal-led derogation. Derogation is a result of negotiations of the key concepts of the Directive 2009/29/EC of the Euro-pean Parliament and of the Council of 23 April 2009 amending Directive 2003/87/EC so as to improve and extend the greenhouse gas emission allowance trading scheme of the Community (further referred to as the “EU-ETS Directive”) carried out by the Polish government. As a result Poland received a concession in the area of mandatory purchase of all emission allowances by power plants via the auction system starting in 2013. According to Article 10c of the EU-ETS Directive some member states, among them Poland, may give a transitional free allocation to installations for electricity production in operation by 31 December 2008 or to installations for electricity produc-tion for which the investment process was physically initiated by the same date. Thanks to the derogation awarded to Poland the plants which existed on 31 December 2008 will only have to purchase some of the allowances they need by auctioning. In 2013 it will be 30% of the average emission in the 2005-2007 period – which is a reference level – or based on weighted emission indicators for different fuels. Then, between 2014 and 2019 the pool of free allowances will be gradually decreased, so in 2020 the full auction system is reached12.

7 ETS – Emissions Trading System. Sectors included in the ETS are electricity generation, production of metals, cement, ceramic materials, glass, pulp and paper, coking plants and refineries.8 E.g. transport, construction industry, services, smaller industrial plants, agriculture, waste management.9 Report from the Commission to the European Parliament and the Council. Progress towards achieving the Kyoto objectives. Brussels, 12 November 2009. COM(2009)630 final, p. 4-5.10 The concept to decrease electricity consumption by 20% was raised during the European Council session held on 23-24 March 2006. The Council has called for the adoption as a matter of urgency of an ambitious and realistic action plan for energy efficiency, bearing in mind the EUenergy saving potential of over 20% by 2020. As a result in October 2006 the European Commission presented the “Action Plan for Energy Efficiency:Realising the Potential”, Communication from the Commission, Brussels, 19.10.2006. COM(2006)545 final.11 See. M.M. Sokołowski, W stronę polskiej polityki klimatyczno-energetycznej. Polska polityka energetyczna – wczoraj, dzisiaj, jutro, Warsaw 2010, p. 67–69.12 M.M. Sokołowski, Społeczne wsparcie rządu, czyli głos ekspertów w kwestii redukcji emisji, Nowa Energia 3/2010, s. 11, also ibidem.

Power Grid Development in Poland in the Context of EU Climate and Energy Package

90

The derogation rules require development of a national plan for infrastructure modernization and impro-vement, as well as clean technologies development13. In Polish conditions the role of such a plan will be played by the National Greenhouse Gas Emission Reduction Programme (or more precisely, its part on investments). Recently, government circles have raised a proposal of changing the plan’s title (for example into a Low-Emis-sion Economy Development Programme).

4. POLISH CONDITIONS AND ACTIONS

The already mentioned targets of emission reduction in the non-ETS sector and increase of RES share in final energy balance are translated into country-specific targets for each EU member state. The only exception isthe energy efficiency commitment, which does not set any compulsory target values. For Poland there are threecountry-specific requirements.

First of all Poland, like any other EU member state – in order to meet European commitments on greenho-use gas emission reduction by 2020 – shall restrict emissions of those gases (in the non-ETS sector) at least by the percentage value determined for this member state in Annex II of the non-ETS Decision14. The target value for Poland is 14% of the 2005 emission.

Secondly, the Polish national general target for the share of renewable sources in final gross energy con-sumption in 2020, according to the stipulations of the Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC, is 15%. When compared to the share of elec-tricity generated by renewable technologies in the final gross energy consumption in Poland in 2005 this meansmore than a twofold growth (in 2005 the share was 7.2%)15.

Thirdly, the presented derogation can only be applied if a national plan that provides for investments in retrofitting and upgrading of the infrastructure and clean technologies is developed.

Two substantial targets, and the third one of a combined character (formal and substantial) require si-gnificant effort, also in the areas of analyses and planning. In appreciation of the significance and scope ofthe required work, the Polish government decided to support the actions of the national administration by an independent public organization. This resulted in the foundation of the Public Advisory Board for the National Emission Reduction Programme (Polish: Społeczna Rada ds. Narodowego Programu Redukcji Emisji) in October 2009 (further referred to as the “Board”).

The Board, with almost 170 members, is an expert entity which formally advises the Minister of Economy. Members of the Board are prominent experts from all branches which have direct or indirect influence on im-plementation of the targets specified by the Climate and Energy Package. They work organized into 17 workinggroups16, divided according to the tasks related to the issue of greenhouse gas emission reduction17. The mission of the Board is to prepare analytical materials for the National Greenhouse Gas Emission Reduction Programme. Those materials include:

• The Green Paper – a document which identifies primary obstacles and problems which interfere withgreenhouse gas emission reduction in Poland

• The White Paper – a document which proposes solutions and concepts for preparation and implemen-tation of the greenhouse gas emission reduction programme in Poland

13 According to Article 10c, Section 5 c), in relation to Article 10c, Section 1, paragraph two of the EU-ETS Directive application of the derogation depends on submitting to the European Commission a national plan that provides for investments in retrofitting and upgrading of the infrastructureand clean technologies.14 According to the Article 3, Section 1 of the Decision 406/2009/EC of the European Parliament and of the Council of 23 April 2009 on the effort of Member States to reduce their greenhouse gas emissions to meet the Community’s greenhouse gas emission reduction commitments up to 2020.15 Maciej M. Sokołowski, W stronę polskiej polityki…, op. cit., p. 69.16 Working Group on Legal Regulations, Working Group on Power Technology for Baseload Sources, Working Group on Construction, Working Group on Economy, Working Group on Nuclear Fuel and Safety, Working Group on Climate and Energy Package, Working Group on Science and Education, Working Group on Strategic Communication, Working Group on Civic Society, Working Group on Organization and Management, Working Group on Environmental Footprint, Working Group on Energy Efficiency, Working Group on Renewable Energy Sources, Working Group on Clean Coal Technolo-gies, Working Group on Grids, Working Group on Market and Working Group on Transport.17 M.M. Sokołowski, Społeczne wsparcie rządu…, op. cit., p. 12.


91

• Assumptions and rules for the National Greenhouse Gas Emission Reduction Programme – position of the Board which includes the key rules to be used during development of the National Programme (like the subsidiarity principle, market primacy principle)

• Road Maps, or in other words, schedules for actions in various areas.The Board also participates in the development process of governmental acts (like the draft of the Energy

Efficiency Act) by presenting its non-binding opinions, supports the government in evaluation of expert reports(like the McKinsey curve report) and carries out international activity (like foreign study visits of the Board’s Secretary General or organizing the visit of Connie Hedegaard, European Commissioner for Climate Action in Poland). Members of the Board actively participate in numerous seminars, debates and conferences, presenting the position of the Board and discussing issues related to greenhouse gas emission reduction18.

5. GRID INVESTMENTS IN POLAND

The Polish commitment to reduce greenhouse gas emissions, develop the renewable energy sector and create an investment plan have one common element – the power grids. This is so because the grid develop-ment is a condicio sine qua non for meeting all those targets resulting from European legal regulations.

Unlocking or making effective use of the RES potential which exists in Poland will require intensificationof actual power grid development. In particular it should focus on constructing a grid of distribution lines al-lowing to connect new generation capacity in eastern and north-eastern areas of Poland. This will additionally allow indentifying “theoretical” investors and separate them from those who really wish to be connected to the system. Optimization of RES usage by effective connection of new sources also increases security of supplies in the country. Despite depreciation of the infrastructure, such sources, which are constructed in much shorter periods than traditional baseload power stations, will allow maintaining reliability of the national power grid in locations where the system is flawed because of its age. In future, however, this unreliability can result withblackouts and outages.

Except for the benefits of introducing RES into the national power grid mentioned above, renewables canalso adversely affect that system. The negative impact is mainly an effect of natural energy carriers’ limitations – first of all wind (and to some extent the sun; however, using solar sources will have a rather minimal impacton the Polish system). This results in a necessity to stabilize the grid operation, so the variation of voltages, as well as active and reactive power do not affect operational reliability.

Grid development which enables RES connections will therefore require further development of systems and equipment assuring cooperation of renewable sources and baseload plants within the power system. Ful-filling European commitments in the RES area also helps to meet a specific target of greenhouse gas emissionreduction. In this way RES get a unique “double” right to be prioritized in development plans created by the government, DSOs and the TSO.

In this context it is also necessary to mention the need to connect nuclear power plants to the Polish grid. Sources of this type, due their low-emission character, fit well in the range of power technologies which helpto meet the pro-reduction commitments. Actual connection, however, requires construction of appropriately strong transmission lines (e.g. dual-circuit 400 kV lines with a total capacity exceeding 3200 MVA). Those lines will be of strategic importance for the grid development and security of supplies of high-quality electricity. The same comments apply to the necessary construction of the so-called Northern Bus – a transmission line running parallel to the Polish coastline, which will enable construction and connection of off-shore wind farms with a target output of some 5000 MW.

18 K. Żmijewski, M.M. Sokołowski, Efektywnie o energetyce cz. 1, Energia i Budynek, 07(38)/2010, p. 13.


92

6. GRIDS – PROBLEMS AND OBSTACLES

Experts of the Board have identified problems and obstacles for grid extension and modernization, andpresented them in the Green Paper19. According to their findings, Poland has a technically obsolete transmissionsystem, which significantly affects the ability to reduce greenhouse gas emissions, control electricity demandand change the structure of energy sources. This also restricts construction and connection of new power plants, among them zero-emission sources. In most cases, lines and transformers are twenty-thirty years old, therefore new investments in construction of new grid elements and modernization of the existing ones are badly needed. According to the data collected for the document “Poland 2030”20 estimated power losses are approximately 12-15%, and that additionally affects reliability of the entire system.

Therefore, according to the Board experts, development of the transmission grid in general – and closing transmission line rings around main Polish urbanized areas (a significant condition of their economic develop-ment) as well as grid development in north-eastern Poland in particular – are the key problems. Technical ana-lyses made for the Development Plan Draft created by PSE Operator SA (Polish TSO) revealed that one of the regions which urgently need investments increasing reliability of the national grid is the north-eastern part of Poland. Investigation of the current system’s emergency operations showed that a failure of a 400 kV line in the northern or north-eastern part of the system will cause a blackout in that part of the country, spreading further towards western areas. This results from the poorly developed network of 400 kV connections and insufficientsource diversification. This situation is extremely dangerous and requires immediate action21.

Unfortunately, complete findings of the Board show Poland as an island system. Existing cross-bordertransmission lines are so weak that they do not allow significant flows or exchange. According to the availabledata their capacity is now approximately 7% of the generation capacity installed in the country, and should be increased to 15% in 2015, 20% in 2020 and 25% in 2030. When analyzing those ambitious targets it needs to be remembered that not only the power industry, but also consumers need development of both Polish and pan--European grids.

It needs to be pointed out, however, that the capacity available for electricity import by AC lines declared by the TSO is only 100-200 MW. Bearing in mind the maximum possible capacity of the Sweden-Poland cable link, it only gives a total capacity of approx. 700 MW available to support the Polish power grid – only 0.28% of the generation capacity installed in Poland. Capacity available for import operations in December 2010 (Net Transfer Capacity) was 0 MW, and the Transmission Reliability Margin during recent years was 500-700 MW22.

No steps aimed at creating appropriate legal regulations supporting smart grid and smart metering de-velopment have been taken yet. Implementation of the so-called Third Energy Package, particularly Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC, has not been started.

Experts of the Board also point out that functionality of transmission and distribution grids is being to-tally transformed now. New needs increase usage of the power grids. RES connections, remote measurements, electric cars, smart grids – all those factors accelerate grid technology development processes. Zirconium grids and the option of replacing AC lines (on all voltage levels) with DC links using thyristor converters open new possibilities.

At this time advanced projects involving energy technologies in distribution grids related to development of electric vehicles (micro- and pico-grids) are being carried out. Electric cars will require development of a network of public and individual (private) charging points, powered by a distribution grid.

Additional obstacles for grid development are posed by the absence of a clear strategy for transmission grid investments in the area of:

19 The Green Paper was published in October 2010 and is available at: www.rada-npre.pl/index.php?option=com_docman&task=doc_downlo-ad&gid=38&Itemid=20 http://www.polska2030.pl/21 Program rozbudowy Krajowej Sieci Przesyłowej w zakresie połączenia Polska – Litwa, Warszawa 2010, p. 7.22 http://www.pse-operator.pl/index.php?dzid=51


93

• Cross-border links• Internal N-S and E-W lines• Closing N-E and S-W loops• Rings around metropolitan areas• Transmission for new baseload capacity• Connection of nuclear power plant• Connection of large wind farms, particularly off-shore sites.

There is also no clear investment strategy for 110 kV distribution grids in the fields of:• Closing loops• Connecting distributed sources, including RES• Transmitting power from new CHP plants and biogas/biomass power plants

Finally, a clear investment strategy for medium (15 kV) and low (230/400 V) voltage networks. Problems here are:

• Supplying power for new investment sites• Re-electrification of rural areas and small towns• Delayed connection of distributed sources• Absence of support mechanisms for household generation23.

To recapitulate: fostering investments in up-to-date grid infrastructure is a key task for decision makers. The problems with that include permitting procedures for accelerated grid construction (as transmission grid development is a process of many years) and development of bilateral connections in a European transmission grid. Lower requirements for power reserves will also lower emission and cost levels. Unfortunately, we lack solutions for financial support of investments and the legal framework (e.g. right-of-way, using road corridors,access to infrastructure, return on capital as a base for tariff drafting etc.).

7. SOLUTIONS AND CONCEPTS

At this time the Board is at an advanced stage of developing its White Paper. The grid issues are covered in Chapter 2. Scope of investments and Chapter 3. Scope of legislation actions. They are discussed in detail in the following sections: 2.1. Electricity system, 2.1.7. Grids, 2.1.7.1. Cross-border links, 2.1.7.2. Transmission grids, 2.1.7.3. Distribution grids (110 kV), 2.1.7. 4. Distribution grids, 2.1.8. Smart grids, 2.1.8.1. Smart metering, 2.1.8.2. Grid sensoring, 3.2. Act on implementation of strategic investments of key importance for national de-velopment, 3.10. Grid investment legislation, 3.10.1. Public roads act, 3.10.2. Spatial planning and management act, 3.10.3. Fixed property management act, and 3.10. 4. Civil code.

Those parts of the White Paper will be extended with the executive part – the planned Road Map for Grids. This map will be an extension of previous road maps presented in June 2010.

Based on the findings made during development of the Green Paper it can be stated that the grid develop-ment in Poland will require actions aimed at development and implementation of:

• New tariff creation mechanism based on return from investment• New mechanism for connecting distributed sources, including RES• Facilitations in investment process, particularly in the right-of-way area within the energy and road

corridor• Current grid monitoring system with special focus on crisis management (icing in the winter, elonga-

tion in the summer); implementing new technologies in grid design (FACTS, multi-conductor cables) is a must

23 So-called prosumers according to terminology proposed by J. Popczyk.


94

• Improvements of the existing RES support system (green certificates) by the introduction of an in-vestment certificate category (short term) for new investments and optional long term certificatesto support operational activity of this proves necessary; in both cases the right to obtain certificatesshould be time limited and awarded within an auction system24, preferring those operators who require lowest support.

8. CONCLUSIONS

Implementation of the Climate and Energy Package in Poland forces us to investigate the technical infra-structure in a comprehensive manner – while earlier studies had been usually narrow-focused. One of the conc-lusions of the new analysis carried out by the Board is highlighting the key role of power grids in the processes of developing low-emission economy, improving electricity generation and consumption efficiency, and improvingsecurity of energy supplies to end customers.

The final success of proposals will be proven by the application of innovative solutions in:• Technology (smart solutions, sensoring)• Tariff creation process (return from investments)• Regulation (mandatory connection to the grid)• Integration (microgrids, microsources).These will bring new quality into the process of supplying electricity to customers.

24 As mentioned in the Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC (OJ L 176, 15.7.2003., also Article 16, Section 1, Energy Law).


NOTE!

1. ARTICLE ×”

multiplication

2. ABSTRACT

www. actaenergetica.org

acta energetica power engineering qtrly 03/2010

Documents

solar energy

energy deposits

huge energy deposit

reactive power

wind power

renewable technologies

power supply systems

solar system