Efficiency Determination of Converter-Fed Induction Motors: Waiting for the IEC Standard

Transcription

1 Efficiency Determination of Converter-Fed Induction Motors: Waiting for the IEC Standard Aldo Boglietti, Fellow IEEE, Andrea Cavagnino, Senior Member IEEE, Marco Cossale, Alberto Tenconi, Senior Member IEEE and Silvio Vaschetto, Member IEEE Politecnico di Torino, Corso duca degli Abruzzi, Torino, ITALY Abstract - European standard concerning the determination of the rated efficiency of converter-fed induction motors is not yet available if not as a draft version. Waiting for the approved standard IEC , in the paper the draft version is analyzed, together the possibility to use IEC Standard, valid for sinusoidal supply, with PWM inverter supply too. At this purpose, the authors carefully defined the concept of additional harmonic losses introduced by the inverter voltage waveform and tested four Total Enclosed Fan Cooled (TEFC) induction motors. The obtained results have put in evidence the potential applicability of IEC for the determination of the efficiency of inverter fed induction motors. The reported results and discussions could be useful for the Standard Commission members working on the definition of tests and procedures to solve this problem. Keywords: Induction motors, efficiency determination, labeling, sinusoidal supply, PWM supply, stray load losses, additional losses, efficiency standards, IEC standards I. INTRODUCTION In the civil and industrial productive sector, among all the used electric machine topologies, the induction motors are typically those more used, due to their simplicity, robustness, low cost and flexibility, both at constant speed (with grid supply) and variable speed (with inverter supply) [1], [2]. As a consequence, the use of higher efficiency induction motors [3], [4] can drive the society towards to significant energy saving and reduction of the CO 2 emissions [5]-[8]. For this reason, many countries in the world are working for defining new efficiency limits and methodologies to measure and label the rated efficiency [9]-[1]. With respect to the European scenario concerning the 2, 4 and 6 poles induction motors, with a rated power in the range from.75 kw to 375 kw and a rated voltage up to 1 V, the International Electrotechnical Commission (IEC) defined the IE1- Standard efficiency, IE2- High efficiency and IE3- Premium efficiency classes in the IEC (28) standard [11]. Even if the European legislative scenario has been recently summarized in detail in [12], it is important to remark that the European Community regulation CE 64/29 imposes that all the motors introduced in the European market will have to respect the minimum efficiency IE classes following the scheduled agenda reported in Table I [13]. The legislative activities for defining the efficiency determination procedures with inverter supply are in progress. In particular, the first edition of the IEC standard, defining specific test methods for the determination of the extra harmonic losses in motors fed by variable frequency converters, was expected early in 213 [14], but, at the writing time, just a draft version is available [15]. Even if the users of inverter-fed induction motors generally use rules of thumbs to estimate the machine efficiency when the load and the speed change (typically based on data obtained by sinusoidal operations, [16]), the rated efficiency labeling under inverter supply by part of the motor manufacturers is surely an hot topic. For this reason, waiting for the final version of the specific IEC standard, in the paper the authors analyze the draft standard IEC to provide some general comments to support the discussion on the additional harmonic losses with inverter supply. In addition, by means of measurements performed on four TEFC induction motors, the authors want investigate the possibility to use under inverter supply the well-consolidated procedures for the efficiency determination valid for sinusoidal supply. In particular, one of the target of the proposed study is to verify if the Stray Load Losses (SLL) can be still determined using the procedure defined by the Standards CEI-EN [17], when the motor is supplied by an inverter source. TABLE I TIME TABLE FOR APPLICATION OF IEC AS IMPOSED BY THE REGULATION CE 64/29 [13] Phase Starting date Requirements 1 16 June January January 217 All the motors must have a minimum efficiency equal to the IE2 Class. All the motors with rated power between 7.5 kw-375 kw must have a minimum efficiency equal to IE3 class or IE2 class if with inverter supply All the motor with rated power between.75 kw-375 kw must have a minimum efficiency equal to IE3 class or IE2 class if with inverter supply /13/$ IEEE 23

2 II. DEFINITION OF THE STRAY LOAD LOSSES AND ADDITIONAL HARMONIC LOSSES In induction motors fed by PWM converters, the switching effects lead to an increase of the total motor losses with respect to operations with sinusoidal supply. As a consequence, the total additional losses with PWM supply could be split in two contributions having spatial or time nature: the former is mainly associated to the fundamental component, the latter is linked to the time harmonic content of the distorted waveforms. Even if the time harmonics can involve spatial effects in the machine airgap, the previous separation is generally accepted both by the standards and technical specifications [15],[18]. Since the additional losses with sinusoidal supply are called Stray Load Losses (SLL), the authors still use this term definition to identify the additional losses due to the PWM fundamental component only, while the further additional losses introduced in the motor by the inverter switching will be identified as additional harmonic losses [19]. III. SUMMARY OF THE ADOPTED IEC STANDARDS In this study the IEC standards used for the efficiency determination are the IEC [17] and the draft version of the IEC [15], reported in outline hereafter. It is important to remark that when [17] is used to estimate the efficiency with inverter supply, all the prescribed tests (no-load test, temperature test and variable load test) have to be obviously performed with this supply source. A. IEC standard For both sinusoidal and PWM supply the indirect approach (1) to the efficiency determination has been considered, where P 1 is the input power and P T are the total motor losses. (1) The total motor losses can be obtained adding the constant losses (P k ), the load losses (stator and rotor joule losses, P s and P r ) and the stray load losses (P SLL ): (2) The constant losses are measured by means of the wellknow no-load test and they are defined as the no-load input power P minus the no-load stator winding losses P S as shown in (3), where P fw are the friction and windage losses (also called mechanical losses), and P ir are the iron losses. (3) The no-load winding losses can be calculated by (4), where R ll, is the average line-to-line resistance reported at the no-load test temperature. From the thermal point of view, the no-load losses can be also considered stabilized if the noload test is carried out immediately after the load-test, obtaining in this way smaller iron loss values with respect to the no-load test performed in cold conditions [2]. 1.5, (4) The no-load test is performed using a variable voltage U with four points ranging from 125% down to 6% of the machine rated voltage, and three point ranging from 5% down to 2%. The friction and windage losses (P fw ) are defined by the intercept of the straight line to zero voltage of the constant power losses (P k ) plot in function of the voltage squared (U 2 ). This procedure for the friction and windage losses determination with sinusoidal supply seems not be applicable for PWM supply as already discussed by the authors in [21]. It is important to underline that these iron losses must be considered as conventional iron losses because they include, also in the case of sinusoidal supply, the additional no-load losses [21]-[23]. The iron losses at the desired load point have to be evaluated plotting the curve of the P ir = P k - P fw against the voltage U, considering only the test point ranging from 125% down to 6% of the rated voltage. The corresponding iron losses at the desired load point are determined from the obtained curve considering the voltage U r which value is obtained by (5). cos sin (5) where: cos 3 ; sin 1 cos R is the winding resistance, U is the terminal voltage and I is the line current at the considered load point. The voltage value defined by (5) takes into account the resistive voltage drop in the stator winding. It is interesting to observe that the previously described iron loss correction is not considered by the IEEE STd.112-B [24]. The load losses (stator and rotor joule losses P s, P r ) to be used in (2) are evaluated by the rated load temperature test where the machine, suitably loaded and with supply power according to the machine rating, is operated until the thermal equilibrium is achieved (gradient of 2 C per hour). The stator and rotor winding losses are respectively calculated by (6) and (7), where k and s are respectively the stator winding resistance corrective coefficient, and the slip (s) corrective coefficient [17]. This coefficients are used in order to refer the load losses at the reference ambient temperature of 25 C. 1.5 (6) (7) In (6) and (7) the input power P 1, the slip s, the current I and the line-to-line resistance R ll are the values measured during the load test for the considered load point; in the same way, P ir is the value determined by the no-load test considering the desired load point. By the IEC , the additional load losses P SLL are evaluated considering the residual losses P Lr calculated for each load point of the load test curve using (8) and (9). In (9), T is the machine torque and n is the rotor speed (in s -1 ). 231

3 (8) 2 (9) The variable load test consists of six load points, with four points equally spaced in the range 25% - 1% of the rated load (the fourth point has to be the rated load), and two points suitably chosen in the range 1% - 15% of the rated load. The variable load test has to be performed as quickly as possible in order to minimize temperature change in the machine during testing. Anyway, for thermally monitored machines (for example by thermal sensors, such as the tested ones) to increase the number of measurements for the six prescribed loads remaining in the same thermal condition the authors suggest to adopt the procedure described in [2]. Smoothing the residual loss data versus the squared load torque, by means of a linear regression, the validity of the test is checked by the correlation coefficient of the linear regression. If the obtained value is less than.95, the worst point has to be deleted and the computation procedure has to be repeated. If the correlation coefficient is still lower than.95, the test is unsatisfactory and the instrumentation and test errors have to be investigated and corrected before to repeat the test. If the obtained correlation coefficient is strictly higher than.95, the stray load losses P SLL can be calculated by (1): (1) where A is the regression line slope of the residual loss data, and T is the machine torque. This approach to the stray load loss estimation is similar to the IEEE Std. 112-B [24]-[27]. B. IEC (draft version) Under PWM supply, the summation of the losses method prescribed by the IEC requires to test the machine with sinusoidal supply (same fundamental frequency and fundamental supply voltage) in order to compute the total motor losses (2). Then, the total losses with PWM supply have to be computed adding to the sinusoidal total losses the additional harmonic losses. Under the assumption that the additional losses due to a converter supply is generally independent by the load, the standard allows to estimate the additional harmonic losses as the difference between the noload losses P with PWM and sinusoidal supply, performing the two no-load tests with the same fundamental quantities. TABLE II NAMEPLATE DATA OF THE TESTED MOTOR Motor A B C D Power, (kw) Speed, (rpm) Power factor Voltage, (V) Current, (A) Winding temperature, ( C) measured at the end of the rated load temperature test. Figure 1. Test bench for efficiency determination: motor under test and brake (top), PWM inverter (bottom, left), and sinusoidal power supply (bottom, right) IV. EXPERIMENTAL ACTIVITY In order to collect the required data to determine the rated efficiency, four three-phase, TEFC, 4 poles, 5 Hz induction motors have been tested as specified in [17] and [15], for sinusoidal and PWM supply respectively. The nameplate data of the considered machines are listed in Table II. The stator three-phase windings of all the motors are delta connected. Fig. 1 shows the Motor B mounted on the test bench, the PWM inverter and the three-phase sinusoidal power supply used in the test campaign. Please note that the IEC define the reference PWM inverter (2-level voltage source inverter) to be used in the test and the specifications for the instruments. In particular, taking into account the rated power of the tested machines, the standard imposes to use a switching frequency equal to 4 khz, avoiding overmodulation. The fundamental voltage of the output voltage has been regulated changing the modulation index. The bus voltage value has been set up in order to obtain the rated voltage at the motor terminals when the modulation index is equal to one. In order to compare the electrical quantities between sinusoidal and PWM supply, the inverter supply tests were performed measuring the first harmonic voltage, the first harmonic and RMS currents values, and the active power absorbed by the motor under test. The rotor speed and the shaft torque are measured too, in order to determine the efficiency by input-output direct method too. 232

4 Torque [Nm] PWM SIN 2 Speed [rpm] No load current [A] sin cold sin hot pwm cold pwm hot pwm cold fundamental pwm hot fundamental Motor A Motor B Motor C Motor D Figure 2. Torque versus speed characteristic for the Motor C with sinusoidal and PWM supply. An important aspect for the execution of the tests under PWM supply is the accurate definition of the fundamental output frequency (5 Hz in the actual study). In fact, an accurate value of the fundamental frequency is mandatory to compute the rotor slip and then the rotor joule losses when [17] is directly used under PWM supply. This aspect is recognized also by the US standard [28], where rated frequency with a tolerance of ±.5% is requested, while the IEC states, as reference condition, fundamental motor frequency equal to rated motor frequency. With the PWM inverter available for the test it was not so easy respect this requirements, but, after some attempts, good results have been obtained. Just as an example, in Fig.2 it is possible to observe that during the load test the synchronous speed for the Motor C is exactly equal to 15 rpm. A. No-load tests: measurements and results As discussed in Section III, the no-load test is the key test to determine the efficiency under PWM supply. Anyway, since the standard for sinusoidal supply has been also used with the converter supply, the no-load tests have been performed in different conditions. In fact, both with sinusoidal and PWM supply, the no-load tests has been performed in cold condition (when the no-load power input varies by 3 % or less, when measured at two successive 3 min intervals) and hot condition (at the steady-state temperature measured at the end of the rated load temperature test, Table II) [17]. Since it was not possible disconnect the motor under test by the brake at the end of the rated load temperature test (see Fig. 1), the machines were preheated in locked rotor condition at the wanted temperature and then the no-load tests were quickly performed. For the four machines, the no-load currents at the rated voltage are shown in Fig.3. It is possible to observe that the rms current growth between sinusoidal and PWM supply is in percentage limited for all the motors such as the effects of the temperature. As expected, the fundamental component of the no-load current under PWM supply is equal to the value measured with sinusoidal supply Figure 4. Figure 3. No-load current at the rated voltage. Iron losses + mechanical losses [W] sin cold pwm cold pwm cold fundamental sin hot pwm hot pwm hot fundamental Motor A Motor B Motor C Motor D Iron plus the mechanical losses at the rated voltage. Iron losses + mechanical losses [W] y = -4E-16x 4 + 3E-11x 3-8E-7x x + 54 y =.35x + 51 SIN - hot condition PWM - hot condition Squared fundamental back e.m.f. voltage [V 2 ] Figure 5. Motor B: no-load mechanical losses determination under sinusoidal and PWM supply by the constant losses P k. In Fig.4 the constant losses (3) at the rated supply voltage are reported, where the mechanical losses are the sum of the windage and friction losses. With the exception of the inverter-fed motor D case, passing from cold to hot conditions a slight reduction of these losses has been obtained, that can be however welcome to reach the efficiency limit. For this reason, in the following developments only the hot conditions will be considered. 233

5 Iron losses [W] SIN - hot condition PWM - hot condition 5 Fundamental back e.m.f voltage [V] Figure 6. Example of the hot condition for the iron losses: Motor B for Sinusoidal and PWM. TABLE III MECHANICAL LOSSES WITH SINUSOIDAL AND PWM SUPPLY Motor A B C D P mechanical sinusoidal, (W) P mechanical PWM, (W) In order to apply the correction (6) to compute the iron losses at the considered load point during the variable load test, the mechanical losses have to be estimated as previously discussed. The standard procedure generally leads to a not accurate value when applied to the PWM supply, as it can be appreciated in Fig.5. Since there is not a evident and physical reason why the mechanical losses should depend on the voltage supply source, the authors suggest to use an high order polynomial interpolation to fit the measurements with PWM supply in the low voltage range [21]. Applying this fitting technique to the motors under test, the estimated mechanical losses with PWM supply are very close to the ones measured with sinusoidal supply see Table III and Fig.5. Since the measurements with sinusoidal supply are simpler and stable to carry out, in the efficiency determination the mechanical losses estimated with sinusoidal supply will be used. Finally, the iron losses curves for sinusoidal and PWM voltage supply have been determined for the four machine. Just for example, Fig. 6 shows the iron losses in hot condition for the Motor B. V. EFFICIENCY DETERMINATION WITH INVERTER SUPPLY The four motors have been tested in accordance to the IEC both with sinusoidal and PWM supply. The test with sinusoidal supply are reported in this section not only to compare the two supply typologies, but also because they are requested by the IEC A. Stray load loss determination As prescribed by the IEC the variable load test has been performed at the end of the temperature tests in order to compute the SLL, first with sinusoidal supply and then under PWM supply. As proposed in [2], in order to minimize the potential risk of measurement errors during the variable load tests, five measures were done for each load maintaining isothermal conditions and then averaging the readings for the subsequent elaboration. For sake of brevity the SLL results will be presented for the Motor B only; similar results have been obtained for the other three machines too. In Fig.7 the residual losses computed by (8) are reported, while the SLL, estimated by (1), are shown in Fig.8. Looking to this figure, it is possible to observe that the increase of the SLL passing from sinusoidal to PWM supply is around 5-6% for each load point. The same percentage has been found for the Motor C and Motor D, while it results double for the Motor A (see the SLL values at the rated load point reported in Table IV). It should be observed that the Motor A is the smaller one in the tested motor set. The obtained results are very interesting because prove that the additional losses due to the PWM fundamental component are additional losses of spatial nature and they changes with the load. As a consequence, it is realistic that they are comparable with the SLL measured with sinusoidal supply. On the other hand, it is possible to assume that the total amount of the additional harmonic losses due to the PWM waveform don t depend by the load [15], [18], [23] y =.412x R² =.9953 y =.388x R² = Residual losses P Lr [W] PWM SIN Squared load torque [N 2 m 2 ] Figure 7. Motor B: residual losses versus the squared load torque. Stray load losses P SLL [W] PWM SIN Squared load torque [N 2 m 2 ] Figure 8. Motor B: stray load losses versus the squared load torque. 234

6 TABLE IV TEMPERATURE TEST RESULTS WITH SINUSOIDAL AND PWM INVERTER SUPPLY Motor A Motor B Motor C Motor D SIN PWM SIN PWM SIN PWM SIN PWM Ambient temperature, ( C) Winding temperature, ( C) Fundamental supply frequency, (Hz) Rotor speed, (rpm) Line-to-line fundamental voltage, (V) Line current, (A) Adsorbed electric power, (W) Load torque, (Nm) Conventional iron losses, (W) Stator joule losses, (W) Friction losses, (W) (**) Rotor Joule losses, (W) Stray load losses, (W) Efficiency by direct method Efficiency by method Efficiency by method corrected at the test temperature [17] (**) by the no-load test in hot condition with sinusoidal supply [15], [21] B. Efficiency determination The results of the rated load temperature tests and the segregation of the different loss contributions are report in Table IV. In the same table the efficiency values computed in accordance to the different considered methods are reported too. In order to apply the IEC standard, the additional harmonic losses of the machine has been measured as the difference between the no-load losses measured with PWM and sinusoidal supply; their values are reported in Table V in the column P. In the same table the difference between the adsorbed electric power in rated load condition for the two considered supply typologies are included ( P Load ). A part for the motor A, the obtained loss differences are very close, further confirming that the additional harmonic loss contribution doesn t depend by the motor load. The four tested machine efficiency as function of the load torque is reported from Fig.9 up to Fig.12. In these figures, the efficiency measured with the direct input-output method under PWM and sinusoidal supply can be directly compared because the respective variable load tests are performed at the same temperature (see Table IV). In addition, the direct method results can be also reasonably compared with the efficiency computed by the standards because during the lab tests the ambient temperature was not so much different by the prescribed room reference temperature of 25 C. As expected, the direct method motor efficiency with PWM supply is lower that the efficiency with sinusoidal supply of about.5-1 percentage points. The difference rises to 2-3 percentage points at 25% of the rated load. Since the SSL are almost constant, the efficiency detriment is mainly due to the increase of the additional harmonic losses, mainly localized in the conventional iron losses and in the stator and rotor joule losses. TABLE V LOSS DIFFERENCE BETWEEN PWM AND SINUSOIDAL SUPPLY P : NO-LOAD LOSSES DIFFERENCE P LOAD : ADSORBED POWER DIFFERENCE AT RATED LOAD Motor P, (W) P Load, (W) P - P Load, (W) A B C D With respect the rotor joule losses it is important to remark that under inverter operations, they will have an increase for sure, but this increase cannot be easy evaluated because the rotor joule losses are computed using the transmitted airgap power and the slip. This approach is valid with sinusoidal supply only, while with inverter supply it should be used on the first harmonic of the electrical quantities only, because the slip is defined on the magnetic rotating field first harmonic in term of space and time components. With PWM inverter the current time harmonics are typically small [23], as it is well evident comparing the RMS stator currents reported in Table IV, and the same rotor joule losses calculation scheme can be used accepting a small errors, but gaining a great approach simplification. These considerations would not be valid with Six-Step inverters due to the presence of low order time harmonics (5th, 7th, 11th, etc.). In addition, since the additional harmonic losses are not load-depending, the additional harmonic losses in the rotor cage are automatically included in the conventional iron losses measured by the noload test. Once looking at Figs.9-12, it is possible to observe that the standard IEC applied to the case of PWM supply allows to compute efficiency values well comparable with those can be estimated using the draft standard IEC

7 Efficency Efficency SIN Direct.83 PWM Direct.82 PWM PWM Load torque [Nm] Figure 9. Motor A: Efficiency versus the load torque. Efficency SIN Direct PWM Direct PWM PWM Load torque [Nm] Figure 1. Motor B: Efficiency versus the load torque. In particular the agreement among the results is very good for the Motor B, C, and D, while a small discrepancy has been found for the Motor A, due to the loss difference of 49 W shown in Table V. It is also interesting to observe that, for the considered motors and the adopted inverter, the efficiencies with PWM supply determined in accordance the two standards are very close to the efficiency computed through direct measurements, as highlighted by the percentage errors shown in Table VI for variable load operations. Summarizing, the result that the standard valid for sinusoidal supply provides acceptable values also for the inverter supply, when correctly applied, is surely a interesting and welcome aspect. VI. CONCLUSION Starting on January 1th 215, the European Community regulation CE 64/29 mandatory imposes to respect the minimum efficiency class IE2 for inverter-fed induction motors with a rated power in the range 7.5 kw-375 kw. At the date of the writing, no specific standards for the efficiency determination with inverter supply have still officially promulgated SIN Direct.85 PWM Direct PWM PWM Load torque [Nm] Figure 11. Motor C: Efficiency versus the load torque. Efficency SIN Direct PWM Direct PWM PWM Load torque [Nm] Figure 12. Motor D: Efficiency versus the load torque. TABLE VI PERCENTAGE ERROR OF THE EFFICIENCY UNDER PWM SUPPLY (REFERRED TO THE DIRECT METHOD RESULTS) 1 : DIFFERENCE BETWEEN IEC AND DIRECT METHOD 2 : DIFFERENCE BETWEEN IEC AND DIRECT METHOD Motor A Motor B Motor C Motor D Load % % % % % % (**) (**) 31% of the load, due to the minimum applicable load of the test bench. (**) not measured, due to the maximum current limitation of the PWM inverter. For this reason, in this paper the authors analyzed the draft version of a standard IEC devoted to this purpose, reporting in outline the test procedures. Starting by the definition of the additional harmonic losses due to the inverter, the possibility to use the wellconsolidated IEC (valid for sinusoidal supply and quite similar to the IEEE 112 method B) in presence of inverter supply is discussed and approached by means of experimental works on four TEFC industrial motors. 236

8 The reported results confirm the validity of the introduced stray load losses and additional harmonic losses concepts under PWM supply. In addition, for the considered tested motors and the used inverter, the IEC can be reasonably used to determine the machine efficiency under inverter supply too. In a foggy legislative scenario, the presented analysis is surely welcome both to manufactures and costumers having the need to determine the efficiency of inverter-fed machines. REFERENCES [1] H. Li, R.S. Curiac, Designing More Efficient Large Industrial Induction Motors by Utilizing the Advantages of Adjustable-Speed Drives, IEEE Trans. on Industry Applications, Vol.46, No.5, pp , September/October 21. [2] M. Benhaddadi, G. Olivier, B. Dima, Energy savings by means of generalization adjustable speed drive utilization, Canadian Conference on Electrical and Computer Engineering, pp , 27. [3] J. Bacher, F. Waldhart, Efficiency Determination of Standard Asynchronous Machines from Start-Up Data, IEEE-ICEM International Conference on Electrical Machines, pp. 1-6, Rome 21. [4] A.Boglietti, A.Cavagnino, L.Ferraris, M.Lazzari, G.Luparia, No Tooling Cost Process for Induction Motors Energy Efficiency Improvements, Transaction on Industry Applications, Vol.41, No.3, May/June 25, pp [5] H.M. Mzungu, A.B. Sebitosi, M.A. Khan, Comparison of Standards for Determining Losses and Efficiency of Three-Phase Induction Motors, IEEE PES Power Africa 27 Conference and Exposition, pp. 1-6, Johannesburg, South Africa, 16-2 July 27. [6] H.L. Pe, R. Curiac, Motor efficiency, efficiency tolerances and the factors that influences them, Rec. Conf. Industry Application Society, Petroleum and Chemical Industry Conference, pp. 1-6, 21. [7] M. Benhaddadi, G. Olivier, J. Yelle, Premium efficiency motors effectiveness, SPEEDAM21 International Symposium on Power Electronics, Electrical Drives, Automation and Motion, pp [8] M. Benhaddadi, G. Olivier, D. Labrosse, P. Tetrault, Premium efficiency motors and energy saving potential, IEEE-IMDC Electric Machines and Drives Conference, pp , 29. [9] E.B. Agamloh, A Comparison of direct and indirect measurements of induction motor efficiency, IEEE-IEMDC Electric Machines and Drives Conference, pp , 29. [1] E.B. Agamloh, The Partial-Load Efficiency of Induction Motors, IEEE Trans. On Industry Applications, Vol.45, No.1, pp , January/February 29. [11] International Standard IEC (28), Rotating electrical machines - Part 3: Efficiency classes of single speed, three phase, cage induction motor. [12] A.T. De Almeida, F.J.T.E. Ferreira, A. Quintino, Technical and economical considerations on super high-efficiency three-phase motors, Conf. Rec. IEEE/IAS Industrial & Commercial Power Systems Technical Conference (I&CPS), 212, pp [13] Commission Regulation (EC) No 64/29, Implementing Directive 25/32/EC of the European Parliament and of the Council with regard to ecodesign requirements for electric motors, Official Journal of the European Union, L 191, , pp , available on [14] International Electrotechnical Commission (IEC), [15] International Standard IEC , Specific test methods for determining losses and efficiency of converter-fed AC motors, Ed.1.-2/1626/CCDV, Draft, 211. [16] M.J. Melfi, Quantifying the Energy Efficiency of Motors on Inverters, IEEE Industry Applications Magazine, Vol.17, No.6, 211, pp [17] International Standard IEC (211), Rotating electrical machines Part 2-1 Standard methods for determining losses and efficiency from tests (excluding machines for traction vehicles). [18] K. Yamazaki, S. Kuramochi, Additional harmonic losses of induction motors by PWM inverters: Comparison between result of finite element method and IEC/TS 634, Conf. Rec. ICEM 212, pp , 212. [19] A. Boglietti, A. Cavagnino, L. Ferraris, M. Lazzari, Induction Motor Equivalent Circuit Including the Stray Load Losses in the Machine Power Balance, Transactions on Energy Conversion, Vol.23, No.3, September 28, pp [2] A.Boglietti, A.Cavagnino, M.Lazzari, M.Pastorelli, International standards for the Induction Motor Efficiency Evaluation: A Critical Analysis of the Stray-Load Loss Determination, Transaction on Industry Applications, Vol.4, No.5, September/October 24, pp [21] A. Boglietti, R. Bojoi, A. Cavagnino, M. Lazzari, Core Loss Estimation Method for PWM Inverter Fed Induction Motors, Conf. Rec. IEEE-IECON 1, Glendale, Arizona, USA, 7-1 November 21, pp [22] A. Boglietti, R.Bojoi, A. Cavagnino, L. Ferraris, No-Load Operations of Induction Motors under PWM Supply, Conf. Rec. IEEE-ISIE 1, Bari, Italy, 4-7 July 21, pp [23] Z. Gmyrek, A. Boglietti, A. Cavagnino, No-Load loss Separation in Induction Motors Operated by PWM Inverters: Numerical and Experimental approaches, Conf. Rec. IEEE-IECON 11, Melbourne, Australia, 7-1 November 211, pp [24] IEEE Std : IEEE Standard Test Procedure for Polyphase Induction Motors and Generators, IEEE Standard, 24 [25] A.T. De Almeida, F.J.T.E. Ferreira, J.F. Busch, P. Angers, Comparative Analysis of IEEE 112-B and IEC 34-2 Efficiency Testing Standard Using Stray Load Losses Testing in Low-Voltage Three-Phase, Cage Induction Motors, IEEE Trans. on Industry Applications, Vol.38, No.2, pp , March/April 22. [26] A.T. De Almeida, F.J.T.E. Ferreira, J.A.C. Fong, Standards for Efficiency of Electric Motors, IEEE Industry Application Magazine, Vol.17, No.1, pp , 211. [27] E.B. Agamloh, An Evaluation of Induction Machine Stray Load Loss From Collated Test Results, IEEE Trans. Industry Applications, Vol.46, N. 6, 21, pp [28] ANSI/AHRI Standard 211, Performance rating of variable frequency drives, Approved by ANSI on April