whitepaper taro - dnv gl … · failure, maintenance strategies and operational constraints, ram...

SAFER, SMARTER, GREENER

WHITEPAPER

TARO Powering performance forecasting for Power Plants

SAFEGUARDING

LIFE,

PROPERTY

AND THE ENVIRONMENT

Date: May 2016

Prepared by DNV GL - Software

© Copyright DNV GL AS 2016. All rights reserved. No use of the material is allowed without the prior written consent of DNV GL AS.

| WHITEPAPER | TARO | www.dnvgl.com/software Page i

Table of contents

1 MOTIVATION .................................................................................................................. 1

2 INTRODUCTION .............................................................................................................. 2

3 RELIABILITY OF POWER SYSTEMS: DEFINITIONS AND CALCULATIONS ................................. 4

3.1 System Adequacy 4

3.2 System security 8

4 CASE STUDY .................................................................................................................. 9

4.1 Equipment failures 10

4.2 Planned maintenance 11

5 RESULTS AFTER RUNNING THE SIMULATION FOR 1,000 LIFECYCLES .................................. 11

5.1 Performance signature 11

5.2 Shortfall per consumer 12

5.3 Availability and Criticality analysis 14

6 CONCLUSION ............................................................................................................... 15

7 ABOUT THE TARO SOFTWARE ........................................................................................ 16

8 ABOUT THE AUTHOR ..................................................................................................... 16

9 REFERENCES ................................................................................................................ 17

| WHITEPAPER | TARO | www.dnvgl.com/software Page 1

1 MOTIVATION

Power interruptions, such as small shortages or complete power failures, have an impact on society; but

industry also suffers considerably from power cuts. On average, the majority of power supply disruptions

from transmission and distribution grids only endure for a few hours but complete shutdowns are not

unusual in some parts of the world. Both types of disruption directly impact production companies but

also critical infrastructure systems such as telecommunication networks, gas and water supply grids and

hospitals. The consequences of these interruptions for industry are lower productivity, inability to deliver

services, higher maintenance costs and unsafe operating conditions for equipment.

Figure 1: Outage per year and duration 20091

Therefore, power supply plays a decisive role in ensuring societies’ economic output. A large number of

industrial facilities around the world depends on their upstream power grid. In refineries, the power

interruptions can cause extensive environmental, economic and safety-related impacts. For mitigation

typically large investments in local power generation are implemented, especially in remote locations or

to leverage readily available high calorific raw material supplies. But also local backup systems are

present to prevent disruptions at the facility.

High availability of power supply can be supported by a very well-established methodology known as

RAM analysis. RAM analysis takes into account reliability, availability and maintainability factors to

predict system performance based on a combination of simulation techniques with the Monte Carlo

method. RAM analysis also empowers the analyst to evaluate the criticality of systems and to understand

how many outages one should expect of the power supply system. By taking into account equipment

failure, maintenance strategies and operational constraints, RAM analysis helps to optimize productivity

and return on investment.

This whitepaper investigates the challenges of applying RAM analysis for power plants in the context of

electricity supply chain for process industries.

1 Allianz Report - Power trip (Special topic)

http://www.agcs.allianz.com/assets/PDFs/GRD/GRD%20individual%20articles/Power_blackout_risks_article.pdf


2 INTRODUCTION

The availability of reliable electric power supply is fundamental for society both in developed and under-

developed countries. Their challenges differ, but some general examples follow:

In developed regions of the world, such as Europe and the United States, grid infrastructures

have been developed over the years and some sections are now ageing. This, added to a

growing demand combined with more frequent extreme weather events, increases the likelihood

of power blackouts.

In Great Britain the Office of Gas and Electricity Markets (Ofgem) is the regulator, supporting

directly the Gas and Electricity Markets Authority (GEMA). Ofgem issues an annual report on

capacity and reliability levels that assesses the gap between production and demand. Power

plants are exiting from the market, increasing the risk of undersupply. The following paragraph

shows a part of the report issued July 2015. “The Capacity Assessment in 2014 shown an

increasing trend on the risks to security of supply as a result of plant closures. Since we

published our 2014 report, more plant [sic] have exited or announced their intention to exit the

market permanently or temporarily. This has been partly offset by a reduction in peak demand at

the national transmission network level since last year: National Grid believes this is mainly due

to increased contribution from embedded generation (seen as negative demand by National Grid

at the transmission level).

The graphs below present our views of the risks to security of supply over the next three winters.

The left hand graph (total blue range) shows our central view (i.e. the outcomes we consider

most likely) of the expected level of security to be delivered by the normal market alone. The

light blue range indicates the risks implied by the FES (which fall within our central view of risks),

and the dark blue range indicates our view of the risks associated with other central outcomes.

The right hand graph shows how the SBR and DSBR National Grid has procured for this winter

have brought the risks to security of supply within the government’s reliability standard (3 hours

Loss of Load Expectancy) for a wide range of credible outcomes.2”

An example of an under-developed region is Africa’s unreliable power supply. According the

World Bank report (Foster & Steinbuks, 2008) most of the continent’s power companies are

unreliable sources of supply and inefficiently generate capacity. This, added to a deficient

maintenance strategy and unpredictable transmission and distribution, fails to offer adequate

electricity supply.

Independent from the region there are also differences between different industrial sectors. In the oil

and gas industry, small power cuts can cause considerable environmental and economic damage, in

addition to posing risks to workers. Taking a refinery as an example:

the direct economic impact is the inability to deliver production according to contracts – power

cuts will stop production and cause contract loss, leading to revenue loss, payment of penalties

and impacte the company´s image.

the environmental impact is that power interruptions might require burning of hydrocarbons by

the refinery’s flaring unit, emitting large amounts of greenhouse gases. If the amount of

2 https://www.ofgem.gov.uk/sites/default/files/docs/2015/07/electricitysecurityofsupplyreport_final_0.pdf

https://www.ofgem.gov.uk/sites/default/files/docs/2015/07/electricitysecurityofsupplyreport_final_0.pdf


pollutants crosses the environmental limit imposed by the legislation, the company will also be

penalised.3

the impact on safety of workers is that lighting is reduced and the likelihood of incidents is much

higher. In the case of failure of safety systems which would manage the shutdown in case of

power cut (e.g. routing flammable material to the flaring unit), the system failure can lead to

catastrophic events.

Finally, the amount of time taken to reinstate the operations in a refinery, start-up time, is

typically long – units might take up to 24 hours to restart as process conditions are being

brought up to their required level.

A Power Trip report from Allianz4 shows that a power cut of 30 minutes can cause an average financial

loss of US$15,709 for medium and large industrial clients. If the power cut lasts longer, for example

eight hours, the financial loss is nearly US$94,000. However, short blackouts (which occur several times

a year in the US) can add up to an annual estimated economic loss of between US$104 and US$164

billion. For countries under development, the problem is worsening, as national power grids tend to be

very unreliable, causing significant impact to production.

Ensuring continuous power generation depends on a number of parameters but mainly on high

availability of systems. This high availability can be influenced by three factors:

Reliability - equipment failures

Maintainability – repairing an equipment before or after its failure

Operability - operational constraints such as the incapacity of fully compressing gas when losing

a redundant compressor

The methodology of a RAM analysis links these three factors, creating – amongst other results – a

production availability figure and a criticality analysis. This can be used to measure system adequacy

and system security, which are important key performance indicators of power systems. The whitepaper

describes how these factors are combined to estimate the availability of a power system.

3 Flaring event - Torrance ExxonMobil Fire smoke

4 Allianz Report - Power trip (Special topic)

http://www.agcs.allianz.com/assets/PDFs/GRD/GRD%20individual%20articles/Power_blackout_risks_article.pdf


3 RELIABILITY OF POWER SYSTEMS: DEFINITIONS AND

CALCULATIONS

“Reliability” when associated with power system analysis is a measure of the ability of the system to

meet customer requirements for electrical energy. Here, “reliability” is usually divided into two aspects:

system adequacy and system security.

System adequacy relates to the existence of sufficient facilities within the system to satisfy customer

demand. These include the facilities necessary to generate sufficient energy and the associated

transmission and distribution facilities required to transport the energy to the actual customer load

points.

System security, on the other hand, is related to the ability of the system to respond to disturbances

arising in the system, such as dynamic, transient, or voltage instability situations.

To determine system adequacy, the following needs to be included in the analysis process:

1. Generation station reliability

2. Composite generation and transmission reliability

3. Distribution reliability

4. Substation and switch reliability

3.1 System Adequacy

The main challenge when calculating System Adequacy is the requirement of assigning a number of

“probabilistic states” to each generator, used to determine the likelihood of a generator operating at

various output levels. In addition, each one of these systems could potentially have loads that are time-

varying, i.e. any point in time the actual system load is determined.

For small studies, it is possible to determine the system adequacy analytically, as the number of

variations is manageable. The calculation increases its complexity when more generators and loads and

its specific states are defined. Having more generators and loads means that the number of variables

needed to describe the system rapidly expands so that it becomes impossible to solve analytically in a

reasonable amount of time. This problem, where a large number of scenarios have to be modelled, is

well-suited for a Monte Carlo simulation.

In the Monte Carlo method, the sampling simulation technique is used to measure the system state at

each point in time. The system state is described by a random generation of different operating states

which will have a corresponding generation power output, whereas the time points will have a

corresponding power demand. With calculated supply and demand for each point, it is possible to

calculate the value of demand not supplied (DNS) for that particular state. This process is repeated for a

number of lifecycles (iterations) giving, by the end of the simulation process, a number of powerful

metrics:

Loss of load probability (LOLP),

Loss of load expectation (LOLE),

Expected demand not supplied (EDNS), and

Loss of energy expectation (LOEE)


An example of a network is shown below in TARO:

Figure 2: Block Flow diagram describing the links between generators and loads

The Block Flow diagram shows a node for each one of the three generators, one node to sum the power

generation, a node for each one of the three loads and a node to sum the electricity generated.

In this example, the corresponding power output of the generator is calculated randomly by selecting a

pre-defined distribution representing the operating state of each generator. The distributions are defined

below:

- Generator 1: Triangular distribution with a mean rating of 90 MW, minimum rating of 75 MW

and maximum rating of 100 MW.





Each load has also a distribution assigned to it describing to represent the power demand every one hour

over five years. The total demand of the system is calculated by summing all the load demands.

- Load 1: Normal distribution with a mean rating of 90 MW and deviation of 10 MW

- Load 2: Constant demand of 75 MW.

- Load 3: Normal distribution with a mean rating of 90 MW and deviation of 5 MW

With the supply and demand estimated, it is easy to calculate the “demand not supplied” (DNS) and this

can be easily noted using the animation mode.

Take the first cycle of the five-year simulation where it is possible to walk, step-by-step throughout the

timeline. After sampling the production from the generators, the corresponding outputs are:

- Generator 1: 95 MW.



This might represent the generator states as: Generator 1 at 90%, Generator 2 at 80% and Generator 3

at 90%. As aforementioned, the total generation output is the sum of all the three generator outputs,

which in this case is 262 MW.


At the same time that the output of the generators is sampled, the demand is also sampled for each load

node. Assuming that the sampling resulted on:

- Load 1: 90 MW.

- Load 2: 75 MW.

- Load 3: 94 MW

Similar to the total output, the total system demand is the sum of all the load demands; 259 MW. In

order to calculate the DNS, the demand is compared to the generation; since the generation is greater

than the demand, all the demand is supplied and the value of DNS is zero.

Another example is when the generator output is:




This might represent the generator states: Generator 1 at 90%, Generator 2 at 100% and Generator 3

at 70%. Assuming that the sampling resulted on:

- Load 1: 100 MW.

- Load 2: 75 MW.

- Load 3: 72 MW

The total system demand is now 247 MW. In this case, the generation is smaller than the demand, so

there is demand that cannot be supplied. The estimated output is unable to meet the demand – the

shortfall is 28 MW.

Figure 3: Animation mode displaying a shortfall in Load 3


Figure 4: Shortfall in Load 3

This shortfall is displayed in red in the node Load 3 and the shortfall is displayed at the description Load

3.

After running 87,600 iterations, a production histogram can be created describing the system likelihood

of producing at a certain capacity level.

Figure 5: Occurrence data

This graph shows a distribution of the size of shortfall for this configuration and the potential output

defined.

Figure 6: Number of Losses

0

2000

4000

6000

8000

10000

70 -

71

71 -

72

72 -

73

73 -

74

74 -

75

75 -

76

76 -

77

77 -

78

78 -

79

79 -

80

80 -

81

81 -

82

82 -

83

83 -

84

84 -

85

85 -

86

86 -

87

87 -

88

88 -

89

89 -

90

90 -

91

91 -

92

92 -

93

93 -

94

94 -

95

95 -

96

96 -

97

97 -

98

98 -

99

99 -

100

Occurrence Data (Nr. Transitions)

0

100

200

300

400

500

600

700

800

7,9 15,8 23,7 31,6 39,5 47,4 55,3 63,2 71,1 79

Number of Losses


Finally, other key performance indicators could be taken into account such as: Loss of Load Probability

(LOLP) and Expected Demand Not Supplied (EDNS).

LOLP = Number of DNS/Number of iterations

In this case, the number of iterations is 87,600 and the number of ENDS is 3,075 which give a LOLP of

3.51%.

If the LOLP is not acceptable, the analysis could be extended to evaluate what parameters could be

potentially modified to decrease the LOLP.

So far the only parameter analysed is system capacity, but RAM analysis allows the method to

incorporate more aspects, such as the reliability and maintainability of the system. This is described in

more detail in the following section.

3.2 System security

System security is the ability of a power system to respond to disturbances arising within that system5.

The central idea is to assess how the current configuration will cope with unplanned and planned

interruptions, causing unavailability. The Monte Carlo method can also be used to estimate the

unavailability of the system. RAM analysis is based on the Monte Carlo method but also taking into

account the functions of failure data, repair time and maintenance strategies.

RAM analysis is typically used to predict the performance of process systems and to provide a basis for

the optimization of such systems. The nature of a RAM analysis will vary according to the purpose of the

study and the scope of work. However, in the majority of cases, the RAM analysis is used to predict

system availability and identify ways to improve system availability by considering both equipment

failures and maintenance factors. The event steps, commonly, should follow the timeline below:

The benefits of running RAM analysis:

Optimize design configuration, maintenance strategy and operational procedures

Reduce maintenance and save costs while maintaining and/or increasing production levels.

Rank capital investment opportunities and support the decision-making process based on

revenue

A decrease in the duration of unplanned and planned outages.

5 R. Billinton and R.N. Allan. Reliability Evaluation of Power Systems. Plenum press, 2nd edition, 1994.

Mean Time To

Repair

Mean Time To

Failure

End of the system’s life

System Downtime

Start-up of the Unit

System Running

System Downtime

Mean Time To

Failure

Mean Time To

Repair

System Running

System Running

Figure 7: Events described in a timeline


Alignment of maintenance resources based on the criticality of equipment to production revenue.

Accurate forecasts of equipment lifecycle costs that reflect equipment age, lifecycle, and

maintenance effectiveness.

Definition of reliability levels for specific systems. Models can be used to estimate frequency of

failures for a certain system and equipment which can then be benchmarked with the

expectation. If the predicted reliability levels are not as expected, changes to design and

equipment selection can be performed to increase the reliability

Life cycle cost (LCC) analyses. Life cycle cost analyses are used to determine the overall cost of

the oil and gas asset during its entire life. RAM analysis is used to estimate the frequency of

failures and, therefore, estimated maintenance cost.

In conclusion, RAM analysis can used to support the decision making process regarding design

configuration, maintenance strategy and operational policy.

The sampling process is repeated for a large number of iterations which are combined at the end of the

simulation process, resulting on a probabilistic view of how the system is likely to behave. The graph

below shows how the Monte Carlo method averages to a stable value after running a number of cycles.

Figure 8: Rolling average a production efficiency from many simulations

The following section extends on the current example; the following case study investigates the

relationship between capacity, reliability and maintenance repair activities. In addition, it accounts for

the ability to shed consumers of lower priority and meet the demand of high priority (e.g. hospitals).

4 CASE STUDY

This case study describes a power generation system including:

Power import from the grid

Some limited modelling of the distribution system (transformers, breakers, substations, etc.).


Power consumers

The power grid is assumed to be under operation for 20 years.

The power generators comprised of one steam turbine and two gas turbines which supply 15 MW and

190 MW (combined power generation), respectively. The power generation from the gas turbines swings

between 95 MW and 80 MW every four months to describe fluctuation on the power generation which

relates with seasonal demand:

Power generation in the winter 95 MW (November to May)

Power generation in the summer is 80 MW (May to November)

The consumers have specific demands and their priority varies from one to nine – Consumer 1 being first

in priority and Consumer 9 last in priority.

Figure 9: Consumer demand

4.1 Equipment failures

Equipment failures are defined using unscheduled elements. These elements are events which

occurrence can be estimated using a specific statistical distribution. The most common statistical

distributions are exponential, Weibull and normal distributions.

An example of data for a circuit breaker 35/72kV and power transformer is shown below:

Table 1: Reliability data

Equipment Failure type MTTF (years) Repair Type MTTR (hours)

Circuit Breaker 35/72kV

Exponential 364 Constant Repair Time 39.1

Power

Transformer

Exponential 91 Constant Repair Time 85.1

The following section discusses the planned maintenance activities.

10

25

35 40

35 30

40

80

45

0

10

20

30

40

50

60

70

80

90

Consumer1

Consumer2

Consumer3

Consumer4

Consumer5

Consumer6

Consumer7

Consumer8

Consumer9


4.2 Planned maintenance

Planned maintenance is normally defined through scheduled elements. These elements are events with

known frequency of occurrence.

The scheduled maintenance of the gas turbines is described as following:

Combustion inspection: Frequency 1 year, duration 120 hrs

Hot gas path inspection: Frequency 2 years , duration 264 hrs

Major inspection: Frequency 6 years, duration 500 hrs

It is important to note that scheduled activities on both turbines will not coincide, as shown in the

following Gantt chart:

Figure 10: Gantt view of scheduled activity

5 RESULTS AFTER RUNNING THE SIMULATION FOR 1,000

LIFECYCLES

The simulation produces a large number of results that can be used to support the decision-making

process. The following discussion focuses on the performance histogram, shortfall for consumers and

criticality analyses. However, this can be easily be extended to assess the maintainability of the system

from a resourcing perspective plus the financial status of the project accounting for operational

expenditure and revenue.

5.1 Performance signature

The first result to be analysed is the performance signature, which shows the distribution of all the

possible “efficiencies” and likelihood of achieving a certain efficiency level. For this case, there is an 80 %

probability that the system will perform with 99.972 % or lower efficiency. It is also important to notice

that the range of estimates is described by this graph – the lowest expected efficiency for the system is

99.916% and the highest is 99.993%.


Figure 11: Performance Signature

5.2 Shortfall per consumer

The export rate of the power plants is driven by a customer demand that is measured on a continuous

basis. Such rates do not remain constant and suffer large variations within relatively short periods.

Taro can deal with these situations via the use of demand profiles. A demand profile replicates a typical

customer need over a period of time. The power generation from the power plant is driven by the need

to meet the current rate defined by the demand profile.

The following graph describes the number of shortfalls per consumer:

Figure 12: Number of Shortfalls

05

101520253035404550

Nu

mb

er o

f sh

ortf

all

s

Number of Shortfalls


Since Consumer 1 is the highest in priority, its shortfall should be the smallest as shown by the graph.

The number of shortfalls increases as one goes down in the priority list. For example, when comparing

the number of shortfalls of Consumer 1 to Consumer 9, which is the lowest in priority, the number of

shortfalls is much larger for the latter.

In addition, the simulation keeps track of the duration of each shortfall.

Figure 13: Average and Minimum duration of Shortfalls

Consumer 1, Consumer 6 and Consumer 9 show the highest average duration of shortfall, which is close

to a half day. This result is counter-intuitive, as one should expect that Consumer 1, the highest in

priority, would show a small duration of shortfall. The same analysis applies to Consumer 6 when

compared to Consumer 7, 8 and 9. Combining this result with the number of contract losses, one can

conclude that Consumer 1 experienced a small number of contract losses but they were longer when

compared to other consumers.

In addition to the minimum and average, the maximum shortfall duration is also tracked:

Figure 14: Maximum duration of Shortfalls

0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4

0,45

Min. Duration Days

Average Duration Days

0

5

10

15

20

25

Consumer1

Consumer2

Consumer3

Consumer4

Consumer5

Consumer6

Consumer7

Consumer8

Consumer9

Max Duration Days


This graph also aligns with the priority when assessing Consumer 9, but again shows some counter-

intuitive results when evaluating Consumer 1 and Consumer 6.

Furthermore, the efficiency of each consumer is calculated:

Figure 15: Consumer’s efficiency

5.3 Availability and Criticality analysis

Taro will calculate the availability for all units described in the Block Flow diagram. Due to the high

reliability level, most units show 100% availability.

The units showing availability below 100% are displayed in the follow graph:

Figure 16: Availability bar

Analysing this table helps to explain why Consumer 1 and Consumer 6 were appearing with long shortfall.

Consumer 1 and 6 distribution system is presenting a minimum difference when it comes to availability

when compared to other consumers. This explains why the long shortfalls have a comparably low

99,85

99,87

99,89

99,91

99,93

99,95

99,97

99,99

Consumer1

Consumer2

Consumer3

Consumer4

Consumer5

Consumer6

Consumer7

Consumer8

Consumer9

Efficiency

Lost Volume

GasTurbineTrain 2

GasTurbineTrain 1

SteamTurbineSupply

15MW

GridSupply280MW

GasTurbineTrain 1

Distribution

Failures

SteamTurbineDistribu

tionFailures

Consumer 6

Distribu

tionSystem

Consumer 1

distribu

tionsystem

GasTurbineTrain 2

Distribution

Failures

Availability % 95,85 95,89 99,63 99,96 99,98 99,98 99,99 99,99 99,99

95

95,5

96

96,5

97

97,5

98

98,5

99

99,5

100

Eff

icie

ncy (

%)

Availability %


number of occurrences. The shutdown was local to the distribution system which caused the shutdown of

consumer, leading to the shortfall. In reality, this could be representing an older part of the power grid.

The criticality analysis can be used to understand where the unavailability is coming from. For example,

taking Consumer 1 distribution line, unavailability is composed of power transformer and circuit breaker

failure:

Figure 17: Node relative loss%

This is a powerful metric used to identify the critical items for the entire grid.

6 CONCLUSION

The ability to forecast system performance has proven to be of significant benefit in a range of industries.

Capturing the interdependency of the different systems in the power plant also plays an important role

to achieve a precise predicted performance figure.

Figure 18: Performance Signature

0

20

40

60

80

100

Power Transformer Circuit Breaker - 35/72kV

Relative Loss %

Node Rel. Loss %


In addition, the shortfall analysis presents an important tool to ensure a secure supply, taking into

account not only the demand but also the capacity, reliability, maintainability and operability of the

system.

Figure 19: Number of Shortfalls

A RAM study will support decision-making that will optimize production at facilities in a safe and

responsible way.

7 ABOUT THE TARO SOFTWARE

This document describes the features and functionality of Taro, DNV GL’s performance forecasting

software analysis tool. Taro incorporates many complex and interrelated parameters into one single

integrated model:

Units reliability

Units capacity

Operational flexibility

Cost analysis

Maintenance strategy

Taro uses event driven simulation technology to predict and quantify the achievable performance (e.g.

productivity) of your asset over its lifecycle, based on a given configuration. This is achieved by taking

into account all factors that can potentially impact the production. A Taro model can be used to identify

the bottlenecks in the system and to evaluate the merits of alternative design changes; configuration

proposals etc. with the aim of optimizing the system design and prioritizing investment.

8 ABOUT THE AUTHOR

Victor Borges, Senior Product Manager at DNV GL, is a chemical engineer with experience performing

risk and reliability analysis for assets in the oil and gas industry. He is responsible for DNV GL’s world-

leading simulation software packages Maros and Taro.

0

10

20

30

40

50

Number of Shortfalls


9 REFERENCES

C.J. Greiner, J. Solvik1, Yongtao Yang and Tore Langeland, “Risk related to Large Scale Implementation of Wind Power into a Regional Power Transmission System”, ESReDA Conference 2012, 15-16 May 2012, Glasgow, UK Foster, V., & Steinbuks, J. (2008). Paying the Price for Unreliable Power Supplies: In-House Generation

of Electricity by Firms in Africa. USA: The World Bank. L. E. Jones et al, ”Strategies and Decision Support Systems for Integrating Variable Energy Resources in Control Centers for Reliable Grid Operations - Global Best Practices, Examples of Excellence and Lessons Learned”, Alstom Grid Inc., Washington DC, on behalf of the US Dept of Energy, Tech. Rep., Dec. 2011. [Online]. Available: http://energy.gov

Reliability Test System Task Force, “The IEEE Reliability Test System -

R. D. Zimmerman, C. E. Murillo-Sánchez, and R. J. Thomas, "MATPOWER Steady-State Operations, Planning and Analysis Tools for Power Systems Research and Education," Power Systems, IEEE Transactions on, vol. 26, no. 1, pp. 12-19, Feb. 2011. R. Billinton, and R.N. Allan, “Reliability Evaluation of Power Systems”, 2nd ed. New York: Plenum Press, 1996, pp. 68-69

http://energy.gov/


ABOUT DNV GL Driven by our purpose of safeguarding life, property and the environment, DNV GL enables organizations to advance the safety and sustainability of their business. We provide classification and technical assurance along with software and independent expert advisory services to the maritime, oil and gas, and energy industries. We also provide certification services to customers across a wide range of industries. Operating in more than 100 countries, our 16,000 professionals are dedicated to helping our

customers make the world safer, smarter and greener.

SOFTWARE

DNV GL is the world-leading provider of software for a safer, smarter and greener future in the energy, process and maritime industries. Our solutions support a variety of business critical activities including design and engineering, risk assessment, asset integrity and optimization, QHSE, and ship management. Our worldwide presence facilitates a strong customer focus and efficient sharing of industry best practice

and standards.

whitepaper taro - dnv gl … · failure, maintenance strategies and operational constraints, ram...

Documents