Table of contents monday, September 9, 1: 30pm-4: 00pm modular Multi-Level Converters, hvdc, and dc grids I 3

Noise and Vibration Issues in Electric Machines

Thursday, September 22, 10:30AM-12:10PM, Room: 102B, Chair: Junichi Itoh, Ali Bazzi

10:30AM Stator Vibration and Acoustic Noise Analysis of FSPM for a Low-Noise Design [#1586]
Chandan Sikder and Iqbal Husain, NC State University, United States

Noise and vibration in FSPM machines is one of the issues where improvement is needed, especially in topologies that are of segmented stator type. The stator mode frequencies and intensity of acoustic noise generated by magnetic radial force is related to the geometry, configuration and material properties. A method for estimating mode frequencies for segmented stator is proposed. A new pole shaping method is proposed to reduce the effect of undesirable mode frequencies on noise and vibration. The mode frequencies have been calculated using analytical models and verified using structural FEA. A low-noise FSPM design technique is proposed based on pole shaping and structural analysis. The effect of the proposed method on electromagnetic performance of the machine has also been investigated. A prototype FSPM is fabricated and tested based on the proposed design.

10:55AM Current Waveform for Noise Reduction of Switched Reluctance Motor in Magnetically Saturated Condition [#332]
Jihad Furqani, Masachika Kawa, Kyohei Kiyota and Akira Chiba, Tokyo Institute of Technology, Indonesia; Tokyo Institute of Technology, Japan

Permanent magnet motor are widely used in electric vehicle application. Advantages of permanent magnet motors are high efficiency and high torque density as well as low acoustic noise. However, permanent magnet motors have a disadvantage of a requirement of rare-earth materials. Rare-earth materials have cost fluctuation and this becomes major problem in electric and hybrid vehicles for mass production. One of the solutions for this problem is rare-earth-free motor. One of the possible motor types is a switched reluctance motor. Advantages of switched reluctance motors are low manufacturing cost, high reliability, robustness, possible operation at high temperature, and high rotational speed. Switched reluctance motors were considered to have disadvantages of low torque, low power density, low efficiency, high acoustic noise and vibration, and uncommon inverter circuit. A part of the authors have shown that a switched reluctance motor is competitive in torque density, efficiency, and power density compared with the permanent magnet motor employed in the leading hybrid electric vehicles. Acoustic noise and vibration is now the major problem to be solved in switched reluctance motors. In this paper, a novel method is proposed to derive the current waveform reduce noise and vibration of a switched reluctance motor in magnetically saturated region. Principle of noise and vibration reduction is based on reducing the variation of the radial force sum. To realize the minimum variation in the sum of radial forces, radial force expression for each phase should be derived. Radial force expression is approximated with Fourier series. With the proper approximation of radial force, the current waveform is derived to minimize the variation of the radial force sum. The proposed current waveform is consist of DC, fundamental, second, and third harmonic components. Finite element analysis and experiment result are included to show the validity of the proposed method.

11:20AM Torque Ripple Reduction Techniques for Stator DC Winding Excited Vernier Reluctance Machines [#640]
Mengxuan Lin, Ronghai Qu, Jian Li, Shaofeng Jia and Yang Lu, Huazhong University of Science and Technology, China

Stator DC winding Excited Vernier Reluctance Machines (DC-VRMs) are a kind of novel machines, which are investigated more and more owing to absence of expensive permanent magnet material. However, DC-VRMs are found to have large torque ripple if they are not carefully designed. There are a lot of existing methods to reduce the cogging torque in permanent magnet (PM) machines, but the design techniques for DC-VRMs have only been mentioned recently in few papers. In this paper, some new methods are proposed including shifted rotor tooth, stepped rotor skewing with multiple modules, T-shape rotor tooth, various rotor tooth arc, rotor tooth- chamfering, different length of air gap, and these methods are based on the techniques for PM machines to reduce torque ripple. The influence of the proposed methods on torque ripple is examined by using Finite Element Analysis (FEA) and it is found that the torque ripple can be reduced significantly.

11:45AM On the cross coupling effects in structural response of Switched Reluctance Motor Drives [#559]
Shiliang Wang, Lei Gu, Babak Fahimi and Mehdi Moallem, University of Texas at Dallas, United States; Isfahan University of Technology, Iran

Vibration and acoustic noise in switched reluctance motor (SRM) drives have been studied by many researchers. However, most of studies are related to vibration in the stator of SRM caused by radial forces acting on the stator itself. In this paper, a complete vibration analysis for a SRM assembly is conducted which considers the stator and rotor vibrations caused by electromagnetic forces as well as mechanical coupling between stator and rotor. Acceleration in the stator and rotor are computed using impulse response method under hysteresis current excitation. Our investigation illustrates a coupling between stator and rotor and its impact on the overall structural response of the machine.

Electrical Drives for Aerospace and Traction Applications

Thursday, September 22, 10:30AM-12:10PM, Room: 101CD, Chair: Gianmario Pellegrino, Davide Barater

10:30AM Asymmetrical Twelve-Phase Induction Starter/Generator for More Electric Engine in Aircraft [#902]
Radu Bojoi, Sandro Rubino, Andrea Cavagnino and Silvio Vaschetto, Politecnico di Torino, Italy

In this paper, a 12-phase asymmetrical induction machine is proposed as embedded starter/alternator connected to the high-pressure shaft of an open rotor jet engine. The main focus of the paper is on the modeling and control of the multiphase machine in order to verify its operation, including overload capabilities and post fault operations. The machine control uses a direct flux vector control scheme based on a multi-three-phase approach, where each three-phase winding set is independently controlled. In this way, the fault tolerant behavior of the drive system is improved. The proposed solution has been experimentally validated with a multi-modular power converter and a reduced scaled machine prototype (10 kW, 6000 rpm). The overload capability has been verified in generation mode for different speed values and torque up to the 150% the rated value. The sudden transition between healthy and faulty modes is reported for open-phases fault.

10:55AM Axial Position Estimation of Conical Shaped Motor for Green Taxiing Application [#1540]
Sara Roggia, Francesco Cupertino, Michael Galea and Chris Gerada, University of Nottigham, United Kingdom; Politecnico di Bari, Italy

This paper considers the possibility of adopting a conical shaped motor for Green Taxiing application. This topology of motor has been selected in order to obviate the presence of external declutching system interposed between the electric actuator and the wheel. An axial force contributes to move the rotor inside-out of the stator. The axial movement of the rotor can be monitored acting on the magnetizing current. The axial sensor-less position estimation method described hereafter envisages the possibility of evaluating the axial position of the rotor during the engaging and disengaging movement from the wheel. The axial position calculation is dependent on the inductance of the motor. An on-line computation of the position has been implemented through the use of high-frequency injection signals.

11:20AM Closed-form approach for predicting overvoltage transients in cable-fed PWM motor drives for MEA [#609]
Giorgio Pietrini, Davide Barater, Carlo Concari, Michael Galea and Chris Gerada, University of Parma, Italy; The University of Nottingham, United Kingdom

The More Electric Aircraft (MEA) concept has set tight constraints for power density and efficiency of electromechanical actuators in aircraft applications. In order to comply with these high power standards, new wide- bandgap (SiC and GaN) semiconductor devices may be exploited. Unfortunately, the extremely short switching times of these devices can easily trigger high frequency ringing voltage at motor terminal in cable-fed PWM motor drives due to pulse reflection. The resultant overvoltage stresses the insulation of stator windings decreasing the motor's lifespan. The most common solutions involve bulky and heavy passive filters, not suitable for MEA design approach, so the overvoltage suppression remains an open question. This paper explores the influence of pulse rising (and falling) time to the magnitude of motor terminal overvoltage through a detailed closed-form analysis of the problem in order to support electrical drive design optimization.

11:45AM An open problem for More Electrical Aircraft (MEA): how insulation systems of actuators can be qualified? [#445]
Giorgio Pietrini, Davide Barater, Giovanni Franceschini, Paolo Mancinelli and Andrea Cavallini, University of Parma, Italy; University of Bologna, Italy

The concept of More Electric Aircraft (MEA) aims to address the demand for efficiency, reliability and maintainability of today aerospace industry by means of an in-depth electrification of the currently hydraulic, mechanical and pneumatic on-board actuators. The high power density of electric actuators shall decrease substantially the aircraft weight as well as its fuel consumption and environmental impact. However, safety and reliability are primary drivers of this sector. This paper will review the main stress factors affecting the lifespan of insulation materials used in aerospace actuators, concentrating in particular to the influence of the low atmospheric pressure at high altitude and to the effect of wide band-gap power devices' short rise times on voltage stress in windings.

Magnetics II

Thursday, September 22, 10:30AM-12:10PM, Room: 102A, Chair: Shuo Wang, Gerard Hurley

10:30AM High Power Density Impedance Control Network DC-DC Converter Utilizing an Integrated Magnetic Structure [#1428]
Ashish Kumar, Jie Lu, Saad Pervaiz, Alihossein Sepahvand and Khurram Afridi, University of Colorado Boulder, United States

This paper introduces a high-power-density high-efficiency isolated dc-dc converter based on the impedance control network (ICN) resonant converter architecture. The ICN converter maintains very high efficiency by achieving zero voltage switching (ZVS) and near zero current switching (ZCS) across a wide range of input voltages, output voltages and output power. High power density is achieved by combining the three inductors of the ICN converter into a single integrated magnetic structure with two coupled windings. Power losses in this integrated magnetic structure are minimized using a finite element analysis (FEA) based design optimization approach. A prototype 550-W, 1-MHz ICN converter incorporating the integrated magnetic structure, designed to operate over an input voltage range of 36 V to 60 V and an output voltage range of 34 V to 55 V is built and tested. The prototyped ICN converter achieves a power density of 462 W/in3, a peak efficiency of 96.7% and maintains efficiencies above 94.8% across its entire operating range.

10:55AM Time-Domain Homogenization of Litz-Wire Bundles in FE Calculations [#1388]
Korawich Niyomsatian, Jeroen Van den Keybus, Ruth Sabariego and Johan Gyselinck, Triphase, ULB, Belgium; Triphase, Belgium; KU Leuven, Belgium; ULB, Belgium

This paper deals with a time-domain homogenization technique for litz-wire bundles embedded in a finite element (FE) model. An elementary FE model is used to determine dimensionless frequency- and time-domain coefficients regarding the skin and proximity effects in litz-wire bundles. Thanks to these coefficients, litz-wire bundles become homogeneous conductors which are easy to integrate into a FE model of a complete device. The method is validated with the reference solution of the 2-D FE transformer model, which is computed by finely discretizing each conductor. The results agree well with the accurate reference solution.

11:20AM High Frequency Core Coefficient for Transformer Size Selection [#98]
Lukas Mueller and Jonathan Kimball, Missouri University of Science and Technology, United States

Transformer design is critical as the demands on power converter efficiency and power density increase. Initial transformer core shape and size selection can be challenging, in many cases necessitating multiple elaborate design iterations to find a suitable core. A number of core constants have been proposed in the past to assist with initial core selection, however, most of them do not consider high frequency losses in transformer cores which are becoming increasingly important. The Kgfe transformer design method is adapted in this paper to consider high frequency copper losses in transformer designs with bi-directional excitation and two windings. The presented expressions can be used to quickly determine the minimum achievable power losses for a given core size, material and wire diameter.

11:45AM Very High Frequency Integrated Voltage Regulator for Small Portable Devices [#1403]
Dongbin Hou, Fred Lee and Qiang Li, CPES, Virginia Tech, United States

As small portable devices (smartphones, tablets, etc.) becomes lighter, thinner, quicker, and smarter, the voltage regulator for the processor is expected to be efficient, miniaturized, integrated, and placed closer to the processor. In this paper, a concept of very high frequency (tens of MHz) 3D integrated voltage regulator for small portable devices is proposed. Both single-phase and 5-phase integrated inductor with NEC flake magnetic material is designed, fabricated and experimentally tested at 20MHz, featuring simple single-via winding structure, small size, ultra-low profile, ultra-low DCR, air-gap-free magnetic core, and lateral non-uniform flux.

Device Short Circuit Capability

Thursday, September 22, 10:30AM-12:10PM, Room: 202C, Chair: Yi Deng, Ty McNutt

10:30AM Robustness in Short-Circuit Mode: Benchmarking of 600V GaN HEMTs with Power Si and SiC MOSFETs [#791]
Nasser Badawi, Abdullah Eial Awwad and Sibylle Dieckerhoff, Technical University of Berlin, Germany

A short circuit e.g. in a half-bridge converter is a severe and potentially destructive operation condition for a power transistor and needs to be turned off quickly and safely. In order to define strategies how to improve the reliability of a power device, it is necessary to understand the failure dynamics during a short-circuit. In this paper, an experimental study focussing on the short-circuit capability of different types of 600V power transistors based on Si, SiC and GaN is presented. Usually, a 10 microseconds short-circuit withstand time at 400V is required for 600V power transistors. Measurement results show that the investigated Si and SiC MOSFETs can withstand short circuit times up to 13 microseconds at 400V and 150 degree, while the normally-off and cascode GaN devices demonstrate considerably less withstand capability. demonstrate considerably less withstand capability.

10:55AM Investigation on the Short Circuit Safe Operation Area of SiC MOSFET Power Modules [#554]
Paula Diaz Reigosa, Francesco Iannuzzo, Haoze Luo and Frede Blaabjerg, Aalborg university, Denmark; Aalborg University, Denmark

This paper gives a better insight of the short circuit capability of state-of-the-art SiC MOSFET power modules rated at 1.2 kV by highlighting the physical limits under different operating conditions. Two different failure mechanisms have been identified, both reducing the short-circuit capability of SiC power modules in respect to discrete SiC devices. Based on such failure mechanisms, two short circuit criteria (i.e., short circuit currentbased criterion and gate voltage-based criterion) are proposed in order to ensure their robustness under short-circuit conditions. A Safe Operation Area (SCSOA) of the studied SiC MOSFET power modules is formulated based on the two proposed criteria.

11:20AM Short-Circuit Protection of 1200V SiC MOSFET T-type Module in PV Inverter Application [#499]
Yuxiang Shi, Ren Xie, Lu Wang, Yanjun Shi and Hui Li, FSU, United States

A de-sat based short-circuit protection scheme using commercial driver for SiC MOSFETs is presented and experimentally verified on 1200V 3-level T-type SiC MOSFET module in this paper. Response time is very critical for the short circuit protection of SiC MOSFETs due to the limited short circuit withstand time (SCWT). Soft turn-off is required to avoid high voltage spike during the turn-off of the fault current. With the presented circuit, 600 ns response time is realized, and a two stage soft turn-off circuit with gate voltage clamping is implemented. A gate voltage stabilizing circuit without affecting the switching loss is also proposed to prevent false trigger. Since de-sat protection scheme is not applicable for the neutral branch of T-type module due to polarity changes of branch voltage, the short-circuit protection of neutral branch is realized with the half-bridge device protection. Detail circuit design for the 1200 V T-type SiC module in 1 kV PV application is described, and experimental results demonstrate the effectiveness of the circuit.

11:45AM Prediction of Short-Circuit-Related Thermal Stress in Aged IGBT Modules [#1037]
Amir Sajjad Bahman, Francesco Iannuzzo, Christian Uhrenfeldt, Frede Blaabjerg and Stig Munk-Nielsen, Aalborg University, Denmark

In this paper, the thermal stress on bond wires of aged IGBT modules under short-circuit conditions has been studied with respect to different solder delamination levels. To ensure repeatable test conditions, ad-hoc DBC (direct bond copper) samples with delaminated solder layers have been purposely fabricated. The temperature distribution produced by such abnormal conditions has been modelled first by means of FEM simulations and then experimentally validated by means of a non-destructive testing technique including an ultra- fast infrared camera. Results demonstrate a significant imbalance in the surface temperature distribution which confirms the hypothesis that short- circuit events produce significantly uneven stresses on bond wires.

Thursday, September 22, 2:00PM-3:40PM

DFIG Based Wind Power Systems

Thursday, September 22, 2:00PM-3:40PM, Room: 203AB, Chair: Wei Qiao, Behrooz Mirafzal

2:00PM Flexible PCC Voltage Unbalance Compensation Strategy for Autonomous Operation of Parallel DFIGs [#59]
Tao Wang and Heng Nian, Zhejiang University, China

This paper proposes a flexible compensation strategy for parallel-connected doubly fed induction generators (DFIGs) when connected to an unbalanced weak grid. The proposed strategy has two main advantages: 1) the voltage unbalance factor (VUF) at the point of common coupling (PCC) can be continuously controlled, thus the flexible trade-off between the balanced DFIG output current and the balanced PCC voltage can be achieved; 2) the negative sequence current required by the compensation can be accurately shared among the parallel DFIG systems according to their respective operation conditions without the need of real-time communication. The proposed strategy is introduced in detail and simulation results are carried out to verify the proposed compensation strategy.

2:25PM Analysis and Comparison of Super- Synchronous Resonance in Small and Large Scale DFIG System [#187]
Yipeng Song, Frede Blaabjerg and Xiongfei Wang, Aalborg University, Denmark

When connected to a parallel compensated weak grid network, both the small and large power scale Doubly Fed Induction Generator (DFIG) system may suffer high frequency resonance (HFR) due to the impedance interaction between the DFIG system and the parallel compensated weak network. Since the parameters of the small and large scale DFIG systems, including DFIG machine parameters and the LCL filter parameters, may vary between 10 to 100 times, the impedance modeling results of small and large scale DFIG system are quite different. Based on the built impedance modeling results, the HFR in small and large scale DFIG system are theoretically analyzed and compared in this paper with the discussion on the influence of PI controller proportional parameters and the digital control delay on the DFIG system impedance shaping. The experimental validation of small scale DFIG system and the simulation validation of large scale DFIG system are conducted to verify the correctness of the analysis.

2:50PM A Super-synchronous Doubly Fed Induction Generator Option for Wind Turbine Applications [#724]
Kee Shin and Thomas Lipo, ABB, United States; University of Wisconsin - Madison (WEMPEC), United States

Doubly fed induction generators are presently used in the large majority of high power wind turbine applications. The current doubly-fed induction generator technology specifies that the turbine operates sub-synchronously to follow the desired power vs. wind speed profile and to operate super- synchronously only during brief periods where wind gusts exceed the nominal maximum value of thrust. This paper suggests that improved operation is possible if the generator is confined to always operate in the super- synchronous state resulting in improved efficiency and greater output power, and the cost effective power converter arrangement and control scheme for the super-synchronous operation are proposed and investigated in this paper.

3:15PM Fault Diagnosis of Wind Turbine Gearbox Using DFIG Stator Current Analysis [#1289]
Fangzhou Cheng, Chun Wei, Liyan Qu and Wei Qiao, University of Nebraska-Lincoln, United States

Gearbox faults are a leading reliability issue in wind turbines. Generator current-based methods have been successfully used in gearbox fault diagnosis and have shown advantages over the traditional vibration-based techniques in terms of implementation, cost, and reliability. This paper proposed a new generator stator current-based fault diagnostic method for the gearboxes in doubly-fed induction generator (DFIG)-based wind turbines under varying rotating speed conditions. Hilbert transform is first used to demodulate the stator current signal, and then a Vold-Kalman filter is designed to separate the nonstationary fault-related components (called the faulty signal) from the demodulated nonstationary stator current signal. Next, a synchronous resampling algorithm is designed to convert the nonstationary faulty signal to a stationary signal. Finally, the power spectral density (PSD) analysis is applied to the resampled faulty signal for the gearbox fault diagnosis. Experimental results obtained from a DFIG wind turbine drivetrain test rig are provided to verify the effectiveness of the proposed method for gearbox fault diagnosis.