01 increasing induction motor drives efficiency understanding the pitfalls

7/25/2019 01 Increasing Induction Motor Drives Efficiency Understanding the Pitfalls

http://slidepdf.com/reader/full/01-increasing-induction-motor-drives-efficiency-understanding-the-pitfalls 1/19

621.85462-83

621.313.333

Ronnie BELMANSWim DEPREZÖzdemir GÖL

INCREASING INDUCTION MOTOR DRIVES

EFFICIENCY: UNDERSTANDING THE PITFALLS

ABSTRACT Although energy conversion in motor drivensystems has a high efficiency, a small increase in efficiency can leadto significant economical en environmental benefits. Therefore, this paper endeavors to present a comprehensive overview of factorsaffecting the overall efficiency of motor driven systems. After givingan overview of different efficiency measurement techniques asembedded in standards, the paper presents the results of efficiency

tests performed in the laboratory according to a number of selectedstandard methods. It is shown that substantial discrepancies exist inresults. Next, the influence of the shape of the efficiency curve on proper motor choice is elucidated. The paper then discusses theeffect of supply voltage unbalance on the performance of an inductionmachine and shows, on the basis of laboratory tests, that furtherdiscrepancies are introduced into measured efficiency values. A simple example illustrates the importance of variable speed drives(VSDs) in certain applications. To conclude, the efficiencyof convertors used in these VSDs and the possible savings potentialof softstarters is briefly treated.

Ronnie BELMANS, Wim DEPREZ

Katholieke Universiteit LeuvenFaculty of Electrical Engineering (ESAT, div. Electa)

Kasteelpark Arenberg 10, 3001 Heverlee (Leuven), BELGIUMe-mail: [email protected]

e-mail: [email protected]

Özdemir GÖL

University of South Australia, School of Electrical and Information Engineering, Adelaide SA 5000, Australia,

e-mail: [email protected]

PROCEEDINGS OF ELECTROTECHNICAL INSTITUTE, Issue 223, 2005



R. Belmans, W. Deprez, Ö. Göl8

1. INTRODUCTION

The combination of diminishing resources and environmental concernsbased on green-house effect considerations and increasing fossil fuel prices,make that energy efficiency has regained importance in recent years. Accordingto the figures published by the European Commission (“European Energy andTransport: Trends to 2030”), between 2000 and 2030, total energy demand islikely to increase by 29 %. Electricity demand will increase by almost 60 % andconsequently, CO2 emissions are forecasted to increase by 29 % as a result [1].This is in direct contrast with the commitment the European Union made bysigning the Kyoto protocol in 1997, that is to reduce its overall green house gasemissions over the period 2008 to 2012 by 8 % compared with the1990 levels[2]. There is a multitude of actions that can be taken to tackle these problems.Clearly, one of the most important class of measures deals with energy saving.

More than 50% of the electricity consumption in the developed countriesand approximately 65% of the electricity used in industry is consumed byelectrical motors [2-3]. Therefore, the efficiency of motor driven systems is ofmajor importance, especially in the case of induction motors constituting the

bulk usage of electric energy in industrial applications. Their running costs canbe up to hundred times or more the purchase price over their service life.Typically, purchase cost is less than 5 %, of the life cycle cost, withmaintenance accounting for about 5 % of the total. The rest is for the energyconsumed. Thus, it is evident that every percentage of efficiency increaseconstitutes considerable economical and environmental benefits.

Industrial companies and individual users could clearly profit from welldesigned, high efficient motor driven systems; even the fact of beingenvironmental friendly can be a marketing argument. However, in the field these

opportunities are often not properly exploited, due to a lack of knowledge and/ormiscommunication between technical (exploitation) and economical (purchase)managment.

In this context, governments also have an informative, an incentive anda regulatory responsibility. There are numerous examples of national andinternational agreements, incentives and initiatives worldwide. For instance;Thailand has a ‘High Efficiency Motor Program’, Korea has a ‘Green MotorProgramme’ (voluntary agreements overseen by Korea Energy ManagementCorporation (KEMCO)), Australia has several government initiatives, mandatorylabelling and voluntary agreements [4]. In the USA some of the most importantinitiatives are the NEMA (National Electrical Manufacturers Association)



Increasing induction motor drives efficiency: understanding the pitfalls 9

PremiumTM Efficiency Electric Motors Program, the BNM02 voluntaryagreements and the EPAct (Energy Policy Act).

For Europe, this is reflected in different ongoing programmes. Forinstance, ‘The European Motor Challenge Programme’ and the most recentclassifications of industrial a.c. motors on the basis of their certified efficiency,e.g. EFF1, EFF2, EFF3 labels of the CEMEP (the European Committee ofManufacturers of Electrical Machines and Power Electronics) voluntaryagreement [2].

Evidently these efficiency classifications can only work properly in theassumption that efficiency measurement methods perfectly reflect the practicalenergy consumption. Moreover, they should be agreed upon in national and

international standards. A closer examination of standards, however, revealsthat there are major discrepancies between methods proposed by differentstandards. This has serious consequences both in terms of issuing certificatesand credibility of the declared efficiency values for decision making inpurchasing motors. The prescribed methods become all the more difficult toassess for non-ideal operating conditions in a realistic application as opposed tothe ideal test conditions of the standard methods. For instance, standards areconspicuously silent on matters pertaining to unbalanced supply or poor powerquality. Also, issuing certificates on the basis of a single efficiency value is

questionable.This paper endeavours to give a comprehensive overview of factorsinfluencing the efficiency of motor driven systems, wether or not using VSDs.Therefore, it gives an overview of the most prevalent standards on efficiencymeasurement and of the different definitions of voltage unbalance. The effect ofvoltage unbalance on induction motors and their performance is addressed.Some standards concerning voltage unbalance and their consequences oninduction machine performance are discussed. For the sake of completeness,three ways to save energy are treated: energy saving by appropriate motorchoice, by installing a frequency converter or by special soft starters.Throughout the paper, the discussion is backed up by results of efficiencymeasurements conducted in the laboratories of the Electrical EngineeringDepartment at the KULeuven.

2. EFFICIENCY MEASUREMENT

In this section, the most prominent standards for determining the

efficiency of three phase induction motors are briefly discussed. The emphasisis on the difference in efficiency values obtained by using these standards in




assessing the performance of a standard squirrel cage induction motor. Themain causes for these differences will be elucidated. A simple exampleillustrates possible consequences of such discrepancies in efficiency valueswhen used trying to optimise energy efficiency of motor driven systems.

For clarity, it must be mentioned that the term ‘efficiency’ in this paperrefers to ‘energy efficiency’ of induction machines. Depending on theapplication, other criteria for efficiency may exist; e.g. low acoustic noise andvibrations or a high output power to weight ratio. The design considerations formeeting these criteria may not always be consistent with energy efficient designstrategies, as there are better core materials, thinner laminations, improvedinter-lamination insulation, more conducting material (copper, thus heavier), etc.

2.1. Energy Efficiency of Induction Machines

Theoretically, the definition of energy efficiency is very simple:

1Pout Ploss

Pin Pinη = = − (1)

In practice however, a series of different standards, based on (1), lead todifferences in efficiency values of several percent [5-6]. The reason for this isthat there are several possible interpretations of this theoretical notation as isexplained in the following.

The theoretical definition (1) divides the efficiency measurement methodsinto two categories: direct and indirect methods. For induction machines, thismeans that for the direct method the output power has to be determined. Thisnecessitates a torque and a speed measurement. But, efficiency valuesobtained by this method also depend on ambient and motor temperature, which

is not desirable for a transparent efficiency comparison. The second methodallows the correction for these temperature values to a specified ambient andreference motor temperature. This is realized by correcting the individual losscomponents. Yet, the main difference between the standards emerges from theway in which the so-called stray load losses as a part of the overall losses aretreated [5-6].

2.2. Losses in Induction Machines

The losses in a three phase squirrel cage induction motor can be dividedinto five categories. These individual loss components and the methods for




determining their values are now discussed briefly. An extensive discussion ofthese loss components can be found in literature [5-7].

The first four loss components are stator and rotor ohmic (Pstator,RI 2 &

Protor,RI 2) losses, core losses (PFe) and friction and windage losses (P fr,w). The

core and friction and windage losses are obtained from a no-load test. Theohmic losses are determined based on stator resistance, slip and input powermeasurements. As mentioned above, most standards also describe how tocorrect the copper losses for a specified ambient temperature and a referencemotor temperature [8-9].

The fifth loss component, the additional load losses, also known as strayload losses, is defined as:

( ) ( ),addit in out Fe stator rotor fr wP P P P P P P= − − + + + (2)

The stray load losses are caused by the space harmonics of stator androtor and by the leakage flux near the winding ends. In the past, severalmethods have been proposed to measure these additional load losses,including the reverse rotation test at slip 2 or half frequency tests at slip -1 and3. These tests make the losses relatively important in comparison with themechanical power.

However, these methods are not very practical to be used for thedetermination of the efficiency of induction machines. In the following sections,it will be explained how the additional losses are included by the three mostimportant standards considered:

• IEEE Standard 112-1996 [8]• IEC 60034-2 Ed.3 [9]• JEC

2.3. IEEE Standard 112

The IEEE 112 standard defines several methods how best to test electricmotors. Efficiency determination is only part of this standard, although it is animportant one. Some of the key (there is a total of 10) test methods forefficiency are:

• Method A: simple input-output• Method B: input-output with loss segregation (or separation)

• Method C: back to back machine test with separation of losses• Method F: equivalent circuit (model) calculation




The other methods, E, E1, F1, C/F, E/F and E1/F1 are variations ofthese. Method A, which is only recommended for small machines, is a directmethod. Method B is in fact an indirect method, but it uses a direct method toobtain a value for the additional losses. The measuring error is reduced bylinearising the additional losses and correcting them for zero additional losses atzero load. The correlation coefficient of the linear regression should be higherthan 0.9. Method B is the recommended and most popular method for testing ofinduction machines up to 180 kW. Therefore, only Method B will be furtherconsidered in this paper.

2.4. IEC Standard 60034

Just as with the IEEE standard 112, the IEC 60034 standard defines howbest to test electrical motors. The second part of this standard, IEC60034-2,describes how to determine losses and efficiency from tests (excludingmachines for traction purposes). This standard also provides differenttechniques to determine the different loss components, e.g. a breaking test withtorque measurement, a back to back test, etc. Historically, due to difficulties of

torque measurement, the current IEC standard for determining motor efficiency(IEC 60034.2 Ed. 3 (1972)) assumes a standard value for the additional loadlosses at rated load of 0.5 % of the input power, proportional to the currentsquared at lower load levels. Note that IEEE112-E1 sets the additional lossesas 1.8 % of the rated output power for machines between 0.75 kW and 90 kW. An intermediate proposed standard, the IEC 61972, gave two possibilities. Thefirst was a method similar to IEEE112-B, the second attributed a fixed amountto every machine of the same rated output power. However, the latest proposeddraft for the revised IEC 60034-2 (4th edition) recommends that for three phase

induction machines between 1 kW and 150 kW the additional losses should bedetermined by the direct method as in the IEEE112-B standard. The approval ofthis proposal by the committee would be a significant improvement as alreadyillustrated in literature [5-6] and by the measurements below.

2.5. JEC Standard on Induction Motor Efficiency

As far as it could be ascertained by the authors, the Japanese JECstandard 37 still completely neglects the additional load losses.




2.6. Measurement Setup

To illustrate the discrepancies that can exist using different standards,a modern small 1.1 kW standard three phase squirrel cage induction motor wastested in laboratory. The test setup that was used is shown in Fig. 1.

Fig. 1. Measurement setup

The brake is a Vibrometer water-cooled Eddy Current - Powder Brakecombination. The torque transducer (type T30FN of HBM) is used incombination with a KMN913.C measurement amplifier of HBM. Thiscombination allows the output power to be determined with an accuracy betterthan 0.5 %. This brake-torque transducer combination can be used to testmachines up to 3 kW at 3000 rpm. The input power, input voltages and currentsare directly measured using a Voltech PM3000A Power Analyser. The two-wattmeter setup is implemented and no current probes are used. The inputpower accuracy is 0.4 %.

The measured input power data are read in a computer usinga LabVIEW® based data acquisition system. The output power data, torque andspeed, have to be entered manually for the different load points. The furtherprocessing of the data is automated, based on a Microsoft® Excel spreadsheet.

PROCEEDINGS OF ELECTROTECHNICAL INSTITUTE, Issue 223, 2005




2.7. Results

Before the start of measurements, ambient and winding temperature aremeasured. Then, the motor is warmed up under rated load until thermalequilibrium is reached. This is checked by measuring the winding temperatureagain. Next, the load test is performed. Starting at 125 % of rated load, about 20load points are set and measured. Finally, the practical no-load test, withdisconnected brake, is performed. During this test, the induction machine is runas a motor with different values of the stator voltage.

Based on the measurements, the spreadsheet automatically generates

efficiency values in four different ways. The first is based on the direct methodas described by (1). The three other efficiency values are the values accordingto IEEE standard 112 Method B, to the actual IEC 60034-2 standard (with thefactor of 0.5 % for the additional losses) and the JEC respectively. Figure 2shows the efficiency values according to these four methods as a functionof normalised torque.

Fig. 2. Efficiency values of a four pole three phase induction motor according to differentstandards

The difference between the measured efficiency at rated load and thecatalogue value is up to 3.7 % for the IEEE112B based efficiency and 1.32 %




for the IEC (0.5 %) one. The difference between the IEEE and the IECefficiency is 2.38 %. The calculated stray load losses are 2.09 % of the ratedoutput power. This clearly illustrates that the old factor of 0.5 % is a seriousunderestimation and that the factor of 1.8 % of IEEE standard 112 Method E1 ismore realistic. These conclusions are in agreement with findings based onmeasurements of 18 motors of 11 kW, 55 kW and 75 KW rating [5-6].

2.8. Discussion

The above shows that measured efficiency values of induction motorsare not unambiguous: they just depend on the standard used in determining theeffieciency! It is illustrated that the only correct way to determine the efficiencyof an induction motor is by the direct determination of the additional losses: allother methods overestimate the efficiency. This could pose problems whenissuing certificates. For instance, the voluntary agreement of CEMEPconcerning the efficiency labels issues an EFF3 label for 4 pole 1.1 kW motorswith an efficiency below 76.2 %, an EFF2 label if the efficiency is above orequal to 76.2 % and an EFF1 label if the efficiency is above 83.8 %. For the

measured motor, this means that according to IEEE standard 112-B, the motordoes not qualify for an EFF2 label, whereas according to the IEC standard(estimating the stay load losses as 0.5 % of rated output power) the motor getsthe EFF2 label comfortably.

In the context of improving the efficiency of motor driven systems andenergy savings in general, the discussion above is very important. However,there are other aspects that influence overall real life efficiency of motor drivensystems that should be considered.

First, there is the issue of stating the efficiency at partial load situations. It

stands to reason that manufactures declare partial load efficiency at, say at50% and 75% of the rated load at least. Since the shape of the efficiency curves(Fig.2.) can differ from manufacturer to manufacturer and from motor type tomotor type, as demonstrated in [5-6], the use of efficiency labels can bemisleading from the point of view of the motor purchaser (see section 4) andunfair from the motor manufacturers’ point of view. It is not clear how or whetherthis problem should be solved. But possible solutions would be the revision ofthe efficiency labels or the introduction of an extra penalising factor reflectingthe shape of the efficiency curve.

A second issue, the effect of power supply quality on efficiency, isadressed in the next section.




3. EFFECT OF VOLTAGE UNBALANCEON INDUCTION MACHINES

This section discusses the effect of power supply quality on efficiency,more specific the effect of voltage unbalance. Standards for efficiencymeasurements prescribe that the supply voltage should be according to thespecifications of the IEC standard 60034-1 allowing for a certain total harmonicdistortion factor and a maximum unbalance of the supply voltage. However,these conditions do not always correspond to real life situations. It is shown thatunbalanced supply voltages considerably affect induction motor efficiency[10-12].

In a three-phase system, a voltage unbalance is the phenomenon inwhich the rms values of the voltages or the phase angles between consecutivephases are not equal. This can occur due to incomplete transposition of powerlines, uneven distribution of single-phase loads, open delta transformerconnections, blown fuses on three phase capacitor banks and so on. Voltageunbalance can negatively influence the efficiency of three phase inductionmachines and even shorten their service life.

3.1. Definitions and Standards

There are several definitions of voltage unbalance in standards andliterature, [10]:

− NEMA uses the line voltage unbalance rate (LVUR) given by

Max Voltage Deviation from Avg Line Voltage% 100

Avg Line Voltage LVUR = ⋅ (3)

− IEEE defines voltage unbalance as the phase voltage unbalance rate(PVUR) as

Max Voltage Deviation from Avg PhaseVoltage% 100

Avg Phase VoltagePVUR = ⋅

(4)




− IEC defines the voltage unbalance factor (VUF) expressed as

2

1

V% 100

VVUF = ⋅ (5)

In equations (3) to (5) V1 and V2 are the positive and negative sequence

voltages respectively, which can be obtained by symmetrical componenttransformation. This is the only definition which includes information of bothmagnitudes and angles. On account of the complex algebra of thetransformation V1 and V2 are phasor quantities. Yet, often only the magnitudeof the VUF is considered. To avoid the use of complex algebra, it is proposed

[10] to use equation (6) which provides a good approximation to the 'true' IECdefinition of unbalance:

2 2 2abe bce cae82 V +V +V

%voltage unbalance 100Avg Line Voltage

⋅= ⋅ (6)

with Vabe being equal to the difference between the line voltage Vab and theaverage line voltage, etc.

For every condition of supply voltage, these three definitions providedifferent values to characterize the unbalance. The other way around, one valueof e.g. the VUF (magnitude value only), can correspond to different unbalancedsituations. This is misleading, yielding erroneous decisions. Especially the PVURdoes not consider any angular information; this can be understood best by theexample of fig. 3. Both cases give the same PVUR, namely 0 %, although it isclear that the situation on the left represents an unbalanced situation. Particularlyin the case of induction motors, where the neutral point, if available, never shouldbe connected, the PVUR should not be used.

Fig. 2. Phasor diagrams illustrating an unbalanced case (left) with only the the phaseangles differing from 120˚ and a balanced case (right). The magnitudes of the phasevoltages are equal in both cases




For induction machines, the line voltages matter since the neutral point ofthe motor shifts with respect to the neutral point of the supply system.Therefore, unbalance characterisation should be based on line voltages only.

Power quality (a criticised term) in general is thoroughly discussed inrecent literature, e.g in [13]. Several IEEE and IEC standards concerning“Power Quality” in fact discuss “normal operating conditions” [13]. Concerningvoltage unbalance, the European voltage characteristics standard, EN50160,states the following [14]:

“Under normal operating conditions, during each period of one week,95% of the 10 minute mean rms values of the negative phase sequencecomponent of the supply shall be within the range 0 to 2 % of the positive phase

sequence component. In some areas with partly single phase or two phaseconnected customers’ installations, unbalances up to about 3 % at three phasesupply terminals occur.”

The ANSI standard limits it to 3 % at the electricity meter under no loadconditions [12]. Such standards are no strict commitment; the supply companiesstrive to deliver a product according to this standard, but they can not or do notalways guarantee it. In practice, the voltage unbalance can exceed the 2 or 3 %level. Moreover, voltage unbalance at the motor terminals can also be causedwithin the infrastructure of companies themselves. Industrial and commercial

facilities may have well balanced incoming supply voltages, but unbalance candevelop within the building due to non uniformly distributed single-phase loads,unbalanced or overloaded equipment, high impedance connections (e.g , bad orloose contacts), badly repaired motors (e.g. when a short on a winding is onlyisolated), etc. Sometimes, unbalance and/or over voltages are also caused byimproper power factor correction. Note also that the standard does not includethe phase angle information.

Voltage unbalances can have detrimental effects on three-phaseinduction motors. This is considered in 3.2. To protect induction machines, theNEMA Standard MG 1-1993: Motors and Generators and the IEC 60034-26prescribe that the machines must be derated when an unbanlance occurs Forinstance, NEMA directs that an unbalance of 3 % requires a 12 % larger motor.It should be noted that these two standards use different voltage unbalancedefinitions.

3.2. Adverse Effect of Voltage Unbalance

The adverse effects of unbalanced voltages on induction motors have

been studied at least since the 1950s [12]. It is common to study the behavior ofthe positive and negative sequence components of the unbalanced supply




voltage to understand the effect of an unbalance on the motor. The positivesequence voltage produces a positive torque, whereas the negative sequencevoltage gives rise to an air gap flux rotating against the forward rotating field,thus generating a detrimental reversing torque. So in fact when neglecting non-linearities, for instance due to saturation, the motor behaves like a superpositionof two separate motors, one running at slip s with terminal voltage V p per phaseand the other running with a slip of (2-s) and a terminal voltage of Vn (Fig. 4).The result is that the net torque and speed are reduced and torque pulsationsand acoustic noise may be registered. Also, due to the low negative sequenceimpedance (R’2/(2-s) !!), the negative sequence voltage gives rise to largenegative sequence currents. At normal operating speeds, the unbalanced

voltages cause the line currents to be unbalanced in the order of 6 to 10 timesthe voltage unbalance [15].

Fig. 4. Graphical representation of the positive andnegative sequence torques of an induction motorsubjected to unbalanced supply vol tages

From Figure 4 it is clear that the entire torque-speed curve is reduced. Inthat context, three points of particular interest on the resulting curve are thestarting, the breakdown and the full load torque. It is clear that the motor takes

longer to speed up in this case. This changes the thermal behaviour of themotor and leads to decreased service life if not early failure. Note that this is




due to the negative torque and/or the reduced positive torque. Moreover, if fullload is still demanded, the motor is forced to operate with a higher slip,increasing rotor losses (R’2/(2-s)) and thus heat dissipation. The reduction ofpeak torques compromises the ability of the motor to ride through dips andsags. Premature failure can only be prevented by derating the machineaccording to standards, allowing it to operate within its thermal limitations.

3.3. Measurements

In order to study how the supply unbalance affects three-phase inductionmotors, the efficiency of an induction motor subject to unbalanced supplyvoltages is measured. The measurement setup is the same as in section 2.6,fig. 1. Also, the same induction motor is used as for the standardised efficiencymeasurements of section 2.6.

To create an unbalanced voltage supply a transformer (230 V : 24 V) isinstalled in each line. The transformers are fed with an adjustable voltage inphase with each corresponding phase voltage of the power supply. Three

different unbalance situations are created one with 2 % VUF and the others withdifferent supply voltage settings with a VUF of 3 %. Table 1 gives an overviewof these different conditions of voltage supply.

TABLE 1Comparison of three unbalanced supply voltage cases in terms of phase voltages of the powersupply, PVUR, LVUR, VUF, the alternative formula for the VUF and the positive and negativesequence components

case Va

[V]

Vb

[V]

Vc

[V]

LVUR

[%]

PVUR

[%]

VUF

[%]

VUFa

[%]

V1

[V]

V2

[V]

3%A 230 230 210 2.96 5.97 2.99 2.98 223.3 ∠ 0 6.67 ∠ -60

3%B 234 234 213 3.06 6.17 3.09 3.07 227 ∠ 0 7 ∠ -60

1% 234 230 218 2.04 4.11 2.11 2.12 227.3 ∠ 0 4.81 ∠ 46.4

The test results are presented in Fig. 5. Efficiency is calculated accordingto the standard IEEE112 Method B. Due to equipment limitations themeasurements with a VUF of 1 % are made up to 95 % of rated load only.




Fig. 5. Efficiency curves of a three phase 1.1 kW induction motor, determined accordingto IEEE standard 112 Method B and for three unbalanced situations

3.4. Discussion

Although only three unbalanced situations are considered, someimportant conclusions can be drawn. Firstly, unbalanced supply conditions thatare well within the margins of the standards can adversely influence theefficiency. In this experiment there is an efficiency decrease of about 1 %maximum for rated load conditions. Note that the IEC definition of voltage

unbalance (VUF) is used. Secondly, the change in efficiency is not solelyproportional to the VUF. This is illustrated by the fact that the efficiency for theVUF of 2 % is worse than the efficiency for the 3 % VUF cases. Thirdly, differentunbalanced supply cases with the same VUF can result in different efficiencyvalues.

3.5. Future Work

The crossing over of the two ‘3 % VUF’ efficiency curves begs for anexplanation which could be found in the difference of positive and negative




sequence voltages, causing the two resulting torque-speed characteristics tohave an intersection. A possible explanation for the worse efficiency of the 2 %unbalance case can be found in the deviation of the angle of the negative andhomopolar sequence voltage. It is not inconceivable that different motor designshave different sensitivity to unbalance, and thus it is possible that an EFF1motor is less efficient than certain EFF2 machines under the same unbalancedconditions. To provide a scientifically based explanation for these effects,further investigation is required using additional measurements on motorsof different rating by different manufacturers (thus of different designs)supported by simulations based on discrete circuit modelling and/or FiniteElement Analysis.

4. ENERGY SAVINGS BY CORRECT DESIGNOF MOTOR DRIVEN SYSTEMS

Whether the motive is directly economical, indirectly to avoid legislativepenalization or environmental, the intention for increasing the efficiency of motordriven systems is energy saving. As discussed earlier, the bulk operational costis energy. Therefore, the pay back times for investments in high efficient motordriven systems are fairly low. In order to properly design these systems andestimate the operating costs, good understanding of total systems and theirefficiency is key. Three important ways to save energy; energy saving by theappropriate motor choice, by a frequency converter or by special softstarters,are discussed in the following.

4.1. Appropriate Motor Choice

As previously mentioned, the purchase price of a motor is relatively small

compared to its energy cost. Therefore, even small increases in efficiency canlead to significant energy savings [5]. As discussed in section 2.8, this does notautomatically mean that simply choosing a motor with a better efficiency (label),so rated values, is the best solution for saving energy. In order to be able tocorrectly predict energy savings, the load profile and the shape of the motor'sefficiency characteristic should be considered.

4.2. Variable Speed Applications

A vast majority of motor driven systems consist of pump, fan, blower orcompressor applications. Historically, these are fixed speed systems using




mechanical mechanisms to control the flow, such as throttle or by-passtechniques or even on-off operation, wasting enormous amounts of energy. Ina variable speed drive, exactly the right amount of energy is delivered to thepump, fan, blower or compressor to obtain the required flow or pressure. Byusing such VSDs, consisting of a standard induction motor and a frequencyconverter, annual energy savings of up to 50 % may be reached [1-2, 5].

Most converters have efficiencies of 95 to 98 %, even at relatively lowloads. It is known that the convertor adversely affects the induction motor'sefficiency. Measurements in the laboratory [5] have shown that the averagedrive efficiency is 2 % lower than the grid connected motor efficiency andbetween different drive combinations, differences in average efficiency up to 4 %

are found. However, this is less important than the energy saving potential.

4.3. Energy Saving by Using Softstarters

For the sake of completeness, the possibility of using a softstarter inorder to save energy, is discussed. The energy saving function of such a deviceis based on the flux optimisation used in the control. At low load conditions, thesupply voltage is lowered in order to decrease the iron losses. This is realised

by phase angle control, using two thyristors in anti-parallel. The fact that no DC-link is required, explains the lower cost when compared to frequencyconverters.

In [5] different softstarters of various power ratings are discussed andtested. It is shown that, as could be expected, the energy saving is maximum atlow loads. Above 50 % load, the increased losses cancel the decreased ironlosses and thus the possible energy saving. A major drawback is the harmonicdistortion of the current leading to additional losses in supply transformers andfaster ageing of the motor. It is also shown that compared to VSDs the payback

times are much higher and the energy savings much smaller.

5. CONCLUSION

This paper gives a comprehensive overview of factors that affect thedetermination of the efficiency of motor driven systems and thus their real life

energy consumption. In doing so, some vagaries in efficiency determination areexposed and discussed. Also the need for the cognizance of operating realities




such as unbalanced supply conditions, which standards presently ignore, isstressed. Finally, some important techniques to save energy in motor drivensystems are presented.

LITERATURE

1. Collard B., Machiels M., Driesen J., Belmans R., Vergels F., Honório L., Laurent M-H.,Zeinhofer H., Evans M-A.: Electric Technologies and their Energy Savings Potential. Unionof the Electricity Industry - Eurelectric, Brussels/Belgium, 2004.

2. De Keulenaer H., Belmans R., Blaustein E. et al.: Energy Efficient Motor Driven Systems.

Can Save Europe 200 Billion kWh of Electricity Consumption and over 100 Million Tonnesof Greenhouse Gas Emissions a Year. European Copper Institute, Brussels/Belgium, 2004.

3. de Almeida A., Bertoldi P., Leonhard W.: Energy Efficiency Improvements in ElectricMotors and Drives. Springer-Verlag, Heidelberg/Germany, 1997.

4. http://www.greenhouse.gov.au/motors/ (5/2005).

5. Slaets B., Van Roy P., Belmans R.: Determining the Efficiency of Induction Machines,Converters and Softstarters. Electromotion 7, no. 2, pp 73-80, 2000.

6. Renier B., Hameyer K., Belmans R.: Comparison of Standards for Determining Efficiencyof Three Phase Induction Motors. IEEE Trans on Energy Conversion 14, no. 3, pp 512-517, 1999

7. Nürnberg W., Hanitsch R.: Die Prüfung elektrischer Maschinen. Springer-Verlag,Berlin/Germany, 1987.

8. IEEE Standard Test Procedure for Polyphase Induction Motors and Generators. IEEE Std112-1996, IEEE Power Engineering Society, New York, NY/USA, 1996.

9. Rotating Electrical Machines – Methods for Determining Losses and Efficiency of RotatingElectrical Machines from Tests. IEC Std 60034-2:1972. CENELEC, Brussels/Belgium,1972.

10. Pillay P., Hofmann P. et al.: Derating of Induction Motors Operating With a Combinationof Unbalanced Voltages and Over or Undervoltages. IEEE Trans on Energy Conversion17, no. 4, pp 485-491, 2002

11. Wang Y-J.: Analysis of Effects of Three-Phase Voltage Unbalance on Induction Motors

with Emphasis on the Angle of the Complex Voltage Unbalance Factor. IEEE Trans onEnergy Conversion 16, no. 3, pp 270-275, 2001

12. Lee C-Y.: Effects of Unbalanced Voltage on the Operation of a Three-Phase InductionMotor. IEEE Trans on Energy Conversion 14, no. 2, pp 202-208, 1999

13. Bollen M.: Understanding Power Quality Problems. IEEE Press, NY/USA, 2000.

14. Voltage Characteristics of Electricity Supplied by Public Distribution Systems. EuropeanStandard EN50160:1999. CENELEC, Brussels/Belgium, 1999.

15. Rotating Electrical Machines – Effects of Unbalanced Voltages on the Performanceof Three-Phase Induction Motors. IEC Std 60034-26:2002. CENELEC, Brussels/Belgium,2002.

Manuscript submitted 01.07.2005




ZWIĘKSZENIE SPRAWNOŚCI NAPĘDÓWZ SILNIKAMI INDUKCYJNYMI:PROBLEMY Z TYM ZWI Ą ZANE

Ronnie BELMANS, Wim DEPREZ

Özdemir GÖL

STRESZCZENIE Pomimo, ż e przekształ canie energii w sys-temach napędowych z silnikami elektrycznymi zachodzi z wysok ą spraw-ności ą , to jednak nawet niewielkie zwi ększenie sprawności moż e pro-wadzi ć do znacz ą cych oszcz ędności. Artykuł ten miał na celu przed-stawienie cał ościowego przegl ą du czynników wpł ywaj ą cych na sprawność systemów napędowych z silnikami elektrycznymi. Po wykonaniu prze-gl ą du róż nych metod wyznaczania sprawności zawartych w normach przedstawiono wyniki sprawności wyznaczonej w laboratorium zgodniez wybranymi metodami. Okazał o si ę, ż e w wynikach badań wyst ę puj ą

istotne róż nice. Omówiono również wpł yw kształ tu krzywej sprawności naodpowiedni wybór silnika oraz wpł yw zasilania napi ęciem niesy-metrycznym na osi ą gi maszyny indukcyjnej. Omówiono tak ż e sprawność przekształ tników stosowanych w napędach i moż liwe oszcz ędności wy-nikaj ą ce ze stosowania softstartu.

01 increasing induction motor drives efficiency understanding the pitfalls

Documents