Performance evaluation of converter topologies for high ... - IEEE Xplore

0 downloads 0 Views 356KB Size Report
For all the topologies, simulations have been carried out considering both 2-Levels and. 3-Levels converter structures, and different machine arrangements.
Performance evaluation of converter topologies for high speed Starter/Generator in aircraft applications Giovanni Lo Calzo*, Pericle Zanchetta**, Christopher Gerada**, Alessandro Lidozzi*, Marco Degano**, Fabio Crescimbini*, Luca Solero* *Department of Engineering, University Roma Tre, Rome, Italy. Email: [email protected] **Department of Electrical and Electronic Engineering, University of Nottingham, United Kingdom. Email: [email protected] Abstract—This paper deals with the performance assessment of different permanent magnet motor drives topologies for Starter/Generator systems in aircraft applications. The comparison focuses on power electronics topologies, evaluating, by means of simulations performed with Matlab/Simulink and Plexim Plecs, some relevant key points such as efficiency, losses and motor current harmonic distortion. For all the topologies, simulations have been carried out considering both 2-Levels and 3-Levels converter structures, and different machine arrangements. All the power electronics systems data are obtained from off-the-shelf power modules characteristics, both Silicon and Silicon Carbide technology based. Keywords—Starter/Generator; PM drives; Inverter; Silicon Carbide; More electric aircraft

I.

INTRODUCTION

In the majority of aircrafts for civil aviation in active service, turbofan engines main role is to produce the trust needed to fly (commonly called primary power), but also the power required for operating all the onboard equipment (commonly called secondary power). Power generated from fuel combustion is in fact used, besides of trust, mainly for four different tasks: - Electric power generation: obtained by means of electric generator connected to the engine shaft trough complex gearbox systems and used for lighting, onboard equipment, kitchen and other electrical loads. - Mechanical power generation: used to operate oil and fuel pumps. - Hydraulic power generation: used to move ailerons and driving operating surfaces and parts, extract and retract landing gears, brake, open and close doors, and for other actuators. - Pneumatic power generation: obtained by channeling part of the high pressure air directly from the turbofan engine (commonly called “bleed air”) and used for cabin pressurization, wings de-icing and air-conditioning. The “More Electric Aircrafts” (MEA) initiative [1]-[7] aims, among various other objectives, to replace the actual power generation architecture with a new one called “bleed-less”, based on the concept of substituting all non-electric systems, with equivalent electric solutions. For example, the bleed-air taken from the engine and used for cabin heating and pressurizing, is the cause of a relevant

k,(((

reduction in the engine efficiency. For this reason, its complete removal is highly desirable. In the same way the engine starting procedure, based on an onboard compressed air reservoir and a complex ducts system built around the engine, does not represent an effective solution as well as the hydraulic subsystem, constituted by heavy and bulky titanium pipes and high power hydraulic actuators with multiple grades of redundancy. However, replacing all non-electric systems with full electric solutions entails a substantial increase in the electric power requirements. In fact, if for aircrafts like the Airbus A330, the onboard installed generating capabilities is about 300kVA, in the new MEA aircrafts it is more than quadrupled, as in the case of the Boing 787 Dreamliner, equipped with four 250kVA generators and two 225kVA auxiliary power units (APU). These increased power requirements impose a radical transformation in both the onboard power distribution system and the generators design. In particular, the MEA initiative is imposing new trends based on massive simplification of all the generating systems. New generating units are intended to be directly connected to the high pressure engine shaft, without complex gearboxes to keep their speed constant and thus working in a wide speed range. They should work as crank motors for the turbofan engines and have higher output voltages in order to reduce conductors’ size. A DC distribution system will be preferred to avoid skin effects in conductors and facilitate the integration of future energy storage units [8]-[11]. This paper presents a detailed performance assessment and comparison among possible drives topologies for a 100kW150kVA on board power generating unit, constituted by a six poles permanent magnets synchronous generator operating from 7 to 33 krpm, and an AC-DC converter connected to a 540V bus, able to operate the generator also as a starter engine. Some works presenting comparisons among different power electronics converters structures in terms of efficiency, power quality etc, already exist in literature [12]-[14], but they do not deal with the use of these converter topologies for complete starter/generators drive systems. This study represents therefore the basis for a proper selection of the optimal topology of a suitable starter/generator solution. II.

TOPOLOGIES UNDER INVESTIGATION

The proposed comparison takes into account four different drive topologies:



-

Single converter, single 3ph machine drives

-

Paralleled converters, multiple 3ph machines drives

-

Series converters, multiple 3ph machines drives

-

Dual fed open-end windings machine drives

two converters and two machines. MSC-MM drives present limited fault-tolerance capabilities: this is due to the fact that dc-link voltage is normally split across multiple converters, leading to increased single unit voltage when bypassing faulted units. Thus, fault tolerance capability is limited by devices voltage characteristics.

A. Single converter, single machine drives Single converter, single machine (SC-SM) drives represent the most common topology adopted in industry. They show limited or absent fault-tolerance capabilities, but their wide spread use and knowledge make them a solid and economic solution. The standard configuration for SC-SM drives is shown in Fig. 1.

Fig. 3. MSC-MM drive topology structure with two converter and two machines Fig. 1. SC-SM drive topology structure

B. Paralleled converters, multiple 3ph machines drives Multiple paralleled converters, multiple 3ph machines (MPC-MM) drives are directly derived from the SC-SM topology by connecting two or more drives in parallel on the same Dc-Link [15]. Fig. 2 shows an example of the MPC-MM arrangement with two converters and two machines.

As in the case of MPC-MM drives, MSC-MM drives can be operated on single machine, provided the necessary split windings. Similar costs considerations, as for the previous topology, can be done with reference to the increased number of converters. D. Dual fed open-end windings machine drives Dual fed open-end windings machine (DF-OEM) drives can be seen as a variant of the MPC-MM drives; however in this case, two inverter legs separately feed each phase of the machine. Fig. 4 depicts the drive’s arrangement.

Fig. 4. DF-OEM drive topology structure

MPC-MM drives present various advantages such as faulttolerance capabilities (with reduced output power) up to the last working unit, reduced torque and dc-link ripples (using appropriate interleaving schemes), single machine operation (splitting single machine windings in multiple sub-windings). On the other hand the increased number of power converters will in turn increase system costs and, if interleaving schemes are adopted, this will also require synchronization between the power electronic units.

DF-OEM drives allow for the highest grade of fault-tolerance capability among the topologies taken into account in this comparison. This is due to the fact that also a fault on the machine side can be managed, granting the drive to continue working even if with degraded performance. However, considering a selected dc-link voltage, this topology requires a machine with higher back EMF voltages, being the converter capable to provide a maximum phase voltage double with respect to a standard full-bridge. This drives arrangement, moreover, is well suited for the use of different power devices technology (Si and SiC) and this feature has been fully exploited in the present comparison.

C. Series converters, multiple 3ph machinse drives Multiple series converters, multiple 3ph machines (MSCMM) drives are directly derived from the SC-SM topology by connecting two or more drives in series on the same Dc-Link. Fig. 3 shows an example of the MSC-MM arrangement with

A particular version of DF-OEM drives is the dual fed open-end winding machine with flying capacitor bridge (DFOEM-FB) [19], as shown in Fig. 5. In this topology the flying bridge converter provides all the reactive power required by the machine, while the main bridge is operated at unity power factor, providing only the required active power.

Fig. 2. MPC-MM drive topology structure with two converter and two machines



TABLE I. reports the characteristics of the power electronic devices used in the proposed comparison. All the chosen power modules come in single inverter leg package and the same power rating has been maintained for all the considered topologies. TABLE I.

Fig. 5. DF-OEM-FB drive topology structure CONVERTER TOPOLOGY

Using this scheme, it is possible to split active and reactive power across the two converters and minimize the reactive power exchange across the dc-link. The main drawback of this topology is the limited fault tolerant capability and the relatively complex control algorithm required to maintain the floating capacitor voltage to the desired value. III.

2-Levels 3-Levels NPC 3-Levels TType

COMPARISON CRITERIA

The performance assessment presented in this work covers all the five drive topologies illustrated in the previous section. For each of those topologies, different power electronic converters structures have been taken into account, considering both 2-Levels and 3-Levels arrangements, and both Si and SiC based power modules. All the configurations tested are summarized in the comparison scheme illustrated in Fig. 6.

2-Levels

DEVICES CHARACTERISTICS

TECHNOLOGY 650V 400A IGBT - Silicon 650V 400A IGBT - Silicon 650V 400A IGBT - Silicon 1200V 180A Dmos - Silicon Carbide

PRODUCT NUMBER FF400R07KE4 F3L400R07ME4 F3L400R07PE4 BSM180D12P2C101

PRODUCER Infineon Technologies Infineon Technologies Infineon Technologies Rohm Semiconductor

All IGBT based configurations have been operated at 20 kHz switching frequency. When SiC carbide devices were used, simulations have been performed also at 40 kHz and 60 kHz switching frequencies. The effect of the increased switching frequency has been taken into account at the control algorithm level and regulators parameters have been differently tuned for each topology in order to evaluate on a fair basis all the analyzed cases. The well-known Field Oriented Control (FOC) algorithm scheme, based on PI regulators in synchronous reference frame, has been used. Depending on power converter arrangements or topologies, this scheme has been slightly modified to deal with the different structures capabilities and requirements. For instance, in the topologies making use of 2-Levels power module, with the exception of the dual fed type, a third harmonic injection has been used in the modulating signal [12], while for all the topologies making use of 3-Levels power modules an additional control loop has been added, acting on 0-axis component of the duty cycles, with the task to balance the dc-link midpoint [20]. This arrangement is briefly sketched in Fig. 7.

Fig. 6. Power electronic topologies comparison scheme

For the 3-Levels converter structures, both Neutral Point Clamped (NPC) and Active Neutral Point Clamped (ANPC, also called T-Type) have been considered as in [16]-[18]. In order to limit the VA capability installed, and thus allowing for the use of a reduced set of commercially available power modules, the multiples converters multiples machines topologies have been simulated limiting the number of converters and machines to two.

Fig. 7. Converters control algorithm scheme

For the dual fed topologies, DF-OEM, a different control scheme has been adopted. In this scheme, each phase of the machine is driven by a single phase H-bridge with unipolar modulation, when the same power modules are used on both the bridges. When mixed Si and SiC modules arrangement was



TABLE II.

used, the Si-IGBT bridge was operated with a switching frequency equal to the machine fundamental frequency (i.e. in square wave mode), and the SiC-Dmos bridge at the desired switching frequency (20, 40 and 60 kHz). This scheme is depicted in Fig. 8.

MACHINES CHARACTERISTICS

DRIVE TOPOLOGY

Ld [μH]

Lq [μH]

Rs [mŸ]

ĭ [Vs]

Poles

Test Speed [rpm]

SC-SM

99

99

1.058

0.0364

6

20000

MPC-MM

198

198

2.116

0.0364

6

20000

MSC-MM

49.5

49.5

0.529

0.0182

6

20000

DF-OEM

303.1

303.1

3.240

0.0637

6

20000

DF-OEMFB

125.3

125.3

1.339

0.0410

6

20000

IV.

SIMULATION RESULTS

Simulation tests have been carried out using the software platforms Matlab/Simulink and Plexim PLECS; each device has been modelled through its V/I and turn-on/turn-off characteristics, respectively to evaluate conduction and switching losses. Achieved efficiency results are reported in Fig. 10.

Fig. 8. DF-OEM Control algorithm scheme

100kW 75kW 50kW

SC−SM 2L

For the DF-OEM-FB topology a different control arrangement was required. This was due to the need to split active and reactive power flow between the two bridges and control the charging of the flying capacitor. For this reason, an additional control loop for the floating capacitor voltage has been added to the main control structure, as shown in Fig. 9.

SC−SM NPC SC−SM TT MPC−MM 2L MPC−MM 2L SiC@20kHz MPC−MM 2L SiC@40kHz MPC−MM 2L SiC@60kHz MPC−MM NPC MPC−MM TT MSC−MM 2L MSC−MM NPC

Fig. 9. DF-OEM-FB Control algorithm scheme

Since the focus of this comparison is the power electronics side of the drive, the machine windings arrangement has been reconfigured to adapt to the chosen topology, however still maintaining the same electrical and mechanical parameters. When double 3ph machine topologies were analyzed, the same machine as in the single machine arrangement was chosen; however it was rewinded and its coils were split as required by the drive topology. Machines characteristics for each of the considered topologies are reported in TABLE II. All the simulations were performed operating the machine at 20krpm (i.e. electrical fundamental frequency of 1kHz) and operating the power electronics in generating mode at 50kW, 75kW and 100kW, with a dc-link voltage of ±270V. The total installed dc-link capacitance was 600μF for all the topologies. The choice of the generating mode was dictated by the fact that this is the expected main operating mode for the converter (over 99% of the total working time).

MSC−MM TT DF−OEM 2L DF−OEM NPC DF−OEM TT DF−OEM 2L − Mixed Si+SiC Sic@20kHz DF−OEM 2L −Mixed Si+SiC Sic@40kHz DF−OEM 2L − Mixed Si+SiC Sic@60kHz DF−OEM−FB 2L 94

94.5

95

95.5

96

96.5

97

97.5

98

98.5

99

Efficiency [%]

Fig. 10. Efficiency simulation results in generating mode and at different power levels



Fig. 11 reports simulation results for currents THD, while in Fig. 12 the evaluation of switching and conduction losses is shown. 100kW 75kW 50kW

SC−SM 2L

efficiency, being machine iron and copper losses also related to frequency. Another aspect that may be of interest is the large difference between 3-Levels NPC and T-Type configurations: even if they produce the same output waveforms, their performances in terms of efficiency are quite dissimilar.

SC−SM NPC

100kW − Conduction 75kW − Conduction 50kW − Conduction 100kW − Switching 75kW − Switching 50kW − Switching

SC−SM 2L

SC−SM TT

SC−SM NPC

MPC−MM 2L

SC−SM TT

MPC−MM 2L SiC@20kHz

MPC−MM 2L

MPC−MM 2L SiC@40kHz

MPC−MM 2L SiC@20kHz

MPC−MM 2L SiC@60kHz

MPC−MM 2L SiC@40kHz

MPC−MM NPC

MPC−MM 2L SiC@60kHz

MPC−MM TT

MPC−MM NPC

MSC−MM 2L

MPC−MM TT

MSC−MM NPC

MSC−MM 2L

MSC−MM TT

MSC−MM NPC

DF−OEM 2L

MSC−MM TT

DF−OEM NPC

DF−OEM 2L

DF−OEM TT

DF−OEM NPC

DF−OEM 2L − Mixed Si+SiC Sic@20kHz

DF−OEM TT

DF−OEM 2L −Mixed Si+SiC Sic@40kHz

DF−OEM 2L − Mixed Si+SiC Sic@20kHz

DF−OEM 2L − Mixed Si+SiC Sic@60kHz

DF−OEM 2L − Mixed Si+SiC Sic@40kHz

DF−OEM−FB 2L

DF−OEM 2L − Mixed Si+SiC Sic@60kHz

0

1

2

3

4

5

6

7

8

9

Total Harmonic Distortion [%]

DF−OEM−FB 2L

Fig. 11. Current THD simulation results in generating mode and at different power levels

Comparing the graphs of Fig. 10, Fig. 11 and Fig. 12, it is possible to conclude that the best compromise between efficiency and current THD has been achived by three converters topologies: the MPC-MM with T-Type modules, the MPC-MM with SiC modules operated at 40kHz and the DF-OEM with SiC modules operated at 60kHz. It should be noted that, even if these three topologies present almost the same efficiency and current THD values, they differ for the spectrum of the output frequency, being the first switching harmonic respectively at 20, 40 and 60kHz. This difference should be taken into account when considering the entire drive

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

Losses − kW

Fig. 12. Conduction and Swiching losses simulation results in generating mode and at different power levels

It is obvious that it was possible to achieve switching frequency values out of reach of IGBT technology capabilities for the topologies equipped with SiC-based power modules,. However when SiC modules have been instead used at the same switching frequency of IGBTs, they have shown a consistent reduction in switching losses and a not so high increase in the conduction losses, resulting in an overall better efficiency.



V.

[9]

CONCLUSIONS

In this paper a comparison between power electronics converter topologies for starter/generator aircrafts applications has been illustrated. As many as 19 converters arrangements have been taken into account, including 2-Levels and 3-Levels structures, both silicon and silicon carbide based, connected in both series, parallel or stand-alone across a provided dc-link, and coupled to both standard and open-end windings 3ph machines. Accurate simulation models using off-the-shelf power electronic modules have been developed and used to evaluate efficiency, current THD and both conduction and switching losses for all the considered configurations. Results obtained at different power levels have been compared in order to highlight the topologies presenting the best balance between efficiency and current THD.

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[11]

[12]

[13]

[14]

REFERENCES [1]

[10]

R. I. Jones, "The More Electric Aircraft: the past and the future?" Electrical Machines and Systems for the More Electric Aircraft (Ref. No. 1999/180), IEE Colloquium on, pp.1-4, 1999 C. R. Avery, S.G. Burrow, P.H. Mellor, "Electrical generation and distribution for the more electric aircraft" Universities Power Engineering Conference, UPEC 2007, pp.1007-1012, 4-6 Sept. 2007 A. Emadi, M. Ehsani, "Aircraft power systems: technology, state of the art, and future trends" Aerospace and Electronic Systems Magazine, IEEE , vol.15, no.1, pp.28-32, Jan 2000 R.T. Naayagi, "A review of more electric aircraft technology" Energy Efficient Technologies for Sustainability (ICEETS), 2013 International Conference on, pp.750-753, 10-12 April 2013 J.A. Rosero, J.A.Ortega, E. Aldabas, L. Romeral, "Moving towards a more electric aircraft" Aerospace and Electronic Systems Magazine, IEEE , vol.22, no.3, pp.3-9, March 2007 A. Boglietti, A. Cavagnino, A. Tenconi, S. Vaschetto, "The safety critical electric machines and drives in the more electric aircraft: A survey" Industrial Electronics, IECON '09, 35th Annual Conference of IEEE, pp.2587-2594, 3-5 Nov. 2009 J.S. Cloyd, "Status of the United States Air Force's More Electric Aircraft initiative" Aerospace and Electronic Systems Magazine, IEEE , vol.13, no.4, pp.17-22, Apr 1998 Zhong Chen, Miao Chen, "Active filter solutions with high reliability for more electric aircraft" IECON 2012 - 38th Annual Conference on IEEE Industrial Electronics Society, pp.197-203, 25-28 Oct. 2012

[15]

[16]

[17]

[18]

[19]

[20]



R. D. Telford, S.J. Galloway, Graeme M. Burt, "Evaluating the reliability & availability of more-electric aircraft power systems" Universities Power Engineering Conference, UPEC 2012, pp.1-6, 4-7 Sept. 2012 P. Magne, B. Nahid-Mobarakeh, S. Pierfederici, "Active Stabilization of DC Microgrids Without Remote Sensors for More Electric Aircraft" Industry Applications, IEEE Transactions on, vol.49, no.5, pp.23522360, Sept.-Oct. 2013 P. Wheeler, “The More Electric Aircraft: Why Aerospace needs Power Electronics?” Proc. of the 13th Conf. on Power Electronics and Applications (EPE ’09), pp. 1–30, 2009. M.P. Kazmierkowski, L.G. Franquelo, J. Rodriguez, M.A. Perez, J.I. Leon, "High-Performance Motor Drives," Industrial Electronics Magazine, IEEE , vol.5, no.3, pp.6,26, Sept. 2011 Guanghai Gong, M.L. Heldwein, U. Drofenik, J. Minibock, Kazuaki Mino, J.W. Kolar, "Comparative evaluation of three-phase high-powerfactor AC-DC converter concepts for application in future More Electric Aircraft" Industrial Electronics, IEEE Transactions on, vol.52, no.3, pp.727-737, June 2005 J.A. Weimer, "Electrical power technology for the more electric aircraft" Digital Avionics Systems Conference, 1993 DASC, AIAA/IEEE, pp.445-450, 25-28 Oct 1993 A. Von Jouanne; Haoran Zhang, "A dual-bridge inverter approach to eliminating common mode voltages and bearing and leakage currents" Power Electronics Specialists Conference, 1997. PESC '97, vol.2, pp.1276,1280, 22-27 Jun 1997 M. Schweizer, I. Lizama, T. Friedli, J.W. Kolar, "Comparison of the chip area usage of 2-level and 3-level voltage source converter topologies" IECON 2010 - 36th Annual Conference on IEEE Industrial Electronics Society, pp.391-396, 7-10 Nov. 2010 P. Panagis, F. Stergiopoulos, P. Marabeas, S. Manias, "Comparison of state of the art multilevel inverters" Power Electronics Specialists Conference, PESC 2008 IEEE, pp.4296-4301, 15-19 June 2008 D.A.B. Zambra, C. Rech, J.R. Pinheiro, "Comparison of Neutral-PointClamped, Symmetrical, and Hybrid Asymmetrical Multilevel Inverters" Industrial Electronics, IEEE Transactions on, vol.57, no.7, pp.22972306, July 2010 Junha Kim, Jinhwan Jung, Kwanghee Nam, "Dual-inverter control strategy for high-speed operation of EV induction motors" Industrial Electronics, IEEE Transactions on , vol.51, no.2, pp.312,320, April 2004 I. Araujo-Vargas, A.J. Forsyth, F.J. Chivite-Zabalza, "Capacitor Voltage-Balancing Techniques for a Multipulse Rectifier With Active Injection" Industry Applications, IEEE Transactions on, vol.47, no.1, pp.185-198, Jan.-Feb. 2011