Development of a Switched Reluctance Motor for Automotive Traction ...

16 downloads 0 Views 2MB Size Report
applications (railway, electric buses,…). Permanent ... worldwide electrification of mobility [1]. The topic on .... with the battery system development for the hybrid.

© EVS-25 Shenzhen, China, Nov. 5-9, 2010 The 25th World Battery, Hybrid and Fuel Cell Electric Vehicle Symposium & Exhibition

Development of a Switched Reluctance Motor for Automotive Traction Applications Saphir Faid1, Patrick Debal1, and Steven Bervoets1 1

Punch Powertrain, R&D Department, Schurhovenveld 4 125, BE3800 Sint-Truiden, Belgium E-mail: [email protected]

Abstract— This paper presents an advanced development of a switched reluctance motor/generator for automotive traction applications. The performance, efficiency and peak torque characteristics of switched reluctance machines, combined with its typical robust, low cost construction make this technology an attractive alternative to other motor types typically chosen for traction applications. In this research, a combined optimization was conducted on the design and control for a compact high torque assist motor/generator aimed at application in full hybrid vehicles. From the early design stage, known challenges related to switched reluctance motors such as torque ripple, vibrations and acoustic noise were addressed by a combined approach of innovative mechanical design features and state of the art control of the motor excitation. The motor was manufactured and characterized, with resulting characteristics covering the application requirements and matching simulated performance and efficiency. The motor was integrated into hybrid- and battery electric test vehicles where performance, NVH levels and durability are assessed while demonstrating the potential of this motor for automotive traction applications. Copyright 2010 EVS25. Keywords— Switched Reluctance Motor, Hybrid, EV, torque ripple, acoustic noise

1

Introduction

Punch Powertrain develops hybrid and electric drivetrains for passenger cars. The electric traction motor is a crucial component as it impacts the vehicle’s performance and because the cost of batteries related to range requirements urges for a highly efficient but costeffective drivetrain. For the projects under development, several options for electric traction motors were investigated, namely permanent magnet motors, induction motors and switched reluctance motors (SRM). Induction motors are the most widely used type in industrial applications as well as heavy traction applications (railway, electric buses,…). Permanent magnet synchronous motors offer significantly better efficiency and power density, which has led to increasing popularity in hybrid and electric passenger cars, electric bicycles, scooters,… However, cost and supply concerns regarding the limited reserves of rare earth magnets are a limiting factor for application of permanent magnet motors in a scenario of serious worldwide electrification of mobility [1]. The topic on the most suitable electric machine remains open, and because of some particular advantages, switched reluctance motors may offer an interesting solution for applications requiring a highly performing but cost-effective solution.

2

would not allow total control of the motor. The shown design is an 8/6 configuration with 8 stator poles and 6 rotor poles which is a typical four phase motor. The four phase operation offers the possibility to achieve full torque in each rotor position and allows smoothening of torque ripple at low speeds as demonstrated in this paper.

Unaligned Position

SRM Properties

A switched reluctance motor produces torque purely through interaction of the stator field with rotor saliency. In figure 1 the basic operation of a switched reluctance motor is illustrated. The stator consists of laminated iron with stator poles and windings. The rotor is just laminated iron. By exciting a pair of opposing stator windings, the principle of reluctance will cause a torque to align the rotor poles with the stator poles. The simplest design can be a single phase motor, but this

Aligned Position Figure 1: SRM Operation

Switched reluctance motors boast advantageous properties such as a simple, robust and low-cost construction, lack of permanent magnets, high efficiency in wide speed range and intrinsically safe operation. In case of sudden short-circuit of a sudden open circuit or even short circuit of the motor phases, current will fade out quickly and the motor can keep

© EVS-25 Shenzhen, China, Nov. 5-9, 2010 The 25th World Battery, Hybrid and Fuel Cell Electric Vehicle Symposium & Exhibition

spinning without issues like hazardous voltage or braking torque due to back EMF. The motor could even continue to operate when one or more phases are defect. This intrinsic safety has led to applications such as aerospace (generators, fuelpumps, flap controls,…), and also makes the technology attractive to automotive applications for vehicle safety in case of drive malfunction or an accident. Switched reluctance motors can offer better efficiency when compared to similar induction motors. When compared to permanent magnet motors the peak efficiency of PMSM may reach higher, but this is typically only in a narrow operating range. Switched reluctance typically deliver a relatively high efficiency in a wide speed range and at medium load, which leads to a high average operating efficiency in mixed drive cycles. The lack of permanent magnets allows operating temperatures in excess of 150°C, limited only by the type of insulation. The lack of permanent magnets is a major a cost benefit, while the simple construction enhances reliability while adding to the cost effective package. Nevertheless, switched reluctance motors also pose inherent challenges. The complexity of excitation and design optimization require advanced drive electronics and specific development. Switched reluctance motors carry a reputation of high torque ripple, especially at low speeds, and acoustic noise. Several studies have focused on these aspects [2-5] which has led to a better understanding of these phenomena, although little literature exists on design guidelines and trade-offs. In its current stage of development Punch Powertrain has been able to reduce these downsides to acceptable levels and further improvements are on the drawing board.

3

Motor Specification

Based on vehicle performance simulations for a hybrid vehicle [6], the following specifications were targeted for this electric motor design: Peak Torque: Peak Power: Speed Range: Continuous Power:

200 Nm 30 kW 0-10 000 rpm 15 kW

In order to achieve optimal integration into the hybrid powertrain package developed by Punch Powertrain [5], the following dimensional constraints were set: Diameter: 225 mm Length: 275 mm

4

Design and Optimization

A comparative study of suitable number of phases and poles resulted in the choice for a four phase 8/6 design. This means that the rotor consists of 3 pole pairs, while the stator consists of 8 pole pairs. This configuration matches with the use of four motor phases, which allows full possibility to smoothen torque

ripple by multi-phase excitation. After an iterative design cycle using analytical tools as well as FEA simulation tools, first design was generated. The design is a compromise of maximal mechanical stiffness for acoustic performance, optimal flux linkage for magnetic performance, coil winding volume for minimisation of copper losses and obviously a minimisation of any other losses such as iron loss and windage loss. An important parameter which impacts the shape of the torque-speed graph is the number of turns of the windings. A higher number of turns results in more torque at low speed, while a lower number of turns offers more performance at higher speeds (figure 2). This is also related to the voltage of the DC supply source, in this case the voltage of the battery pack.

Figure 2: Torque-Speed for varying number of coil windings per pole

The supply voltage was determined in accordance with the battery system development for the hybrid powertrain developed by Punch Powertrain, and was set to a nominal voltage of 307V DC which corresponds with the nominal voltage of a series string of 96 LiFePO4 cells which each have a nominal voltage of 3,2V. The simulation proved that it was possible to achieve the required performance within the given sizing constraints, while operating the motor from a 307V DC supply voltage. For more precise simulations, the control parameters are calculated with a state of the art optimization platform. The control parameters that need to be defined for each speed and torque value are: the on-angle, freewheeling angle, off angle and current amplitude. This software tools searches for the optimal set of control parameters. The optimal is a weighted function to have the highest efficiency, lowest noise, lowest torque ripple and lowest iron losses (figure 3). A smoothing algorithm is applied to eliminate sudden deviations in the control parameter tables. Based on simulations in a matrix of operating points, maps could be generated which illustrate properties in the entire torque vs. speed range.

© EVS-25 Shenzhen, China, Nov. 5-9, 2010 The 25th World Battery, Hybrid and Fuel Cell Electric Vehicle Symposium & Exhibition

Figure 4: (From left to right) Completed rotor, stator and assembled motor

5

Characterization

The manufactured prototypes were validated on a test rig with full measurement equipment such as a torque measurement device for accurate characterization (figure 5). The characterization on the test rig allowed more precise determination of the efficiency map and actual torque delivered by the motor.

Figure 5: Motor test rig setup for characterisation

Figure 3: Efficiency and torque ripple under different optimisations

After completing design iterations, the electromagnetic design was determined. The motor design was further detailed according to the hybrid powertrain it matches to. Prototypes were manufactured in order to validate the design (figure 4).

During characterization on the test bench, these calculated control parameters were tested to deliver the requested torque and efficiency. After optimization of the control parameters, the final torque-speed map was determined, the ‘fingerprint’ of the motor from a vehicle point of view. As visible on figure 6, the motor actually delivers higher peak power at medium speeds and achieves efficiency in excess of 90% (motor + inverter) in a wide speed range.

© EVS-25 Shenzhen, China, Nov. 5-9, 2010 The 25th World Battery, Hybrid and Fuel Cell Electric Vehicle Symposium & Exhibition

7

Implementation & In-Vehicle Assessment

After characterization the motor was integrated with the other subsystems into a complete parallel hybrid powertrain as illustrated in figure 8.

Figure 6: Efficiency map (motor + inverter)

6

Torque Ripple Reduction

At low speeds e.g. below 200 rpm, the torque ripple with the conventional control system is prominent that in a traction application the vehicle becomes uncomfortable or even undrivable. Figure 7 shows measurements of actual torque during 360 degrees of rotation, for different levels of requested torque. With the conventional control system there are four prominent torque peaks, corresponding to the excitation of the four stator pairs over one rotor rotation. Current profiling is an adjustment of the amplitude of the requested current in function of the actual rotor position. In the positions with torque peaks, the current will be reduced and vice versa. With the technique of current profiling, this ripple can reduced and virtually eliminated at low and medium torque levels while still delivering the same average torque over one rotation. Only for the maximal torque level, it is not possible to eliminate the ripple without reducing the average delivered torque as the current is already at the maximum level.

Figure 7: Effect of current profiling

Figure 8: Integration to hybrid powertrain through chain drive

The powertrain was then implemented into test vehicles where performance and drivability could be assessed (figure 9). During the first assessment, torque ripple was investigated and brought to an acceptable level for comfortable driving, even at low speeds. The motor performance was fulfilling its expectations delivering the requested torque precisely, and demonstrating reliability. Regarding NVH aspects there were no issues with motor vibrations transmitted to the vehicle, and the acoustic noise produced by the motor itself although clearly perceivable at low speeds especially in generating mode, seemed acceptable to most of the users (subjective assessment). It must me noted that controls parameters obviously have a significant impact on acoustic noise as well, and that this is also part of the overall compromise in optimizing the excitation parameters. Another point to note is the fact that the power electronics unit in this case was also a source of acoustic noise and was perceived as more annoying to the users due to the high pitch noise related to the switching frequency of the IGBT modules. However this is something which can be improved by better sealing of the housing of the power electronics unit, positioning of this unit in the vehicle and by shifting the switching frequencies (although the last aspect is related to switching losses and hence to inverter efficiency).

© EVS-25 Shenzhen, China, Nov. 5-9, 2010 The 25th World Battery, Hybrid and Fuel Cell Electric Vehicle Symposium & Exhibition

improvement. Hence, Punch Powertrain will build on this experience and come up with an even more performing solution.

10 Acknowledgements

Figure 9: Complete hybrid powertrain and test vehicle

8

Further Work

While the motor design has proved to perform according to the requirements and according to simulations, the basic functionality is present. In addition, the issue of torque ripple was eliminated successfully. However, the topic of acoustic noise needs further work which is ongoing at the publication of this paper. In the first place the acoustic noise is measured on a static setup using an acoustic camera to visualize the noise emitted from the motor housing. Secondly, the amplitude of the acoustic noise is measured in a vehicle in different driving conditions, and compared with amplitude and frequencies of acoustic noise emitted by other electric motor types in various applications. Then, an objective statement can be made on the acoustic performance of the presented switched reluctance motor design. Further improvement is still possible, in various ways. On the motor itself, there is still room for improvement regarding stiffness of the stator, for example by insertion of ceramic spacers between stator poles [8]. Apart from the motor design, a lot can be done on acoustic noise handling by application of acoustic insulation materials on the motor or into the vehicle, minimization of structure borne noise transfer paths, and also by calibration of the vehicle in order to eliminate specific noise peaks by modifying the requested combination of torque from electric motor and thermal engine.

9

Conclusions

The conducted motor design and optimization has proven successful with regard to delivered torque and efficiency. The issue of torque ripple, which could be a major drawback for traction applications, was successfully addressed by use of a four phase design and application of current profiling. The integration into test vehicles allowed assessment in the actual application, and first subjective assessments by various test users were positive, although these are subjective results, more objective measurements are to be carried out. The operational test vehicles are nevertheless an illustration of the potential of this low-cost technology. The positive feedback, ideas for even better design and control, and availability of options for noise handling offer further room for

The development of the hybrid powertrain at Punch Powertrain is supported by the Flemish Government as an IWT industrial research and development projects. The IWT is the Institute for the promotion of Innovation by Science and Technology in Flanders.

11 References [1] Oakdene Hollins, Lanthanide Resources and Alternatives, http://www.oakdenehollins.co.uk, March 2010 [2] R.S. Colby et.al. Vibrating modes and acoustic noise in a 4 phase switched reluctance motor. Conference Record of the 1995 IEEE, 1:441–447, 1995. [3] D.E. Cameron et.al. The origin and reduction of acoustic noise in a doubly salient variarable-reluctance machine. IEEE Transactions on Industry Applications, 28:1250–1255, 1992. [4] C. Pollock et.al. Acoustic noise cancellation technique for switched reluctance drives. IEEE Transactions on Industry Applications, 33:477–484, 1997. [5] T.J.E. Miller. Switched reluctance motors and their control. Magna Physics Publishing, New York, 1993. [6] D’hulster F., Stockman K., Belmans R., Modeling of switched reluctance machines: state of the art, International Journal of Modelling and Simulation, Vol. 24, No 4, 2004, pp. 216-223. [7] P. Debal et.al. Development of a Post-Transmission Hybrid Powertrain, EVS24, May 13-16, 2009, Stavanger, Norway [8] Rasmussen et.al. Structural Stator Spacers - The key to silent electrical machines, March-April 2004, IEEE Transactions on Industry Applications, Issue 2, p574 - 581

12 Authors ing. Saphir Faid Punch Powertrain Schurhovenveld 4 125, BE3800 SintTruiden, Belgium Tel: +32 11 679 193 Fax: +32 11 679 230 Email: [email protected] URL: www. punchpowertrain.com Saphir Faid graduated in 2004 as Master in Electro-Mechanical Engineering from GroupT University College in Leuven, Belgium. He worked on several electric vehicle projects including solar powered cars and a fuel cell race vehicle, before joining Punch Powertrain in 2008. Saphir is responsible for subsystems and components of the hybrid powertrain, including the development of switched reluctance motors. ir. Patrick Debal Punch Powertrain Schurhovenveld 4 125, BE3800 SintTruiden, Belgium Tel: +32 11 679 266 Fax: +32 11 679 230

© EVS-25 Shenzhen, China, Nov. 5-9, 2010 The 25th World Battery, Hybrid and Fuel Cell Electric Vehicle Symposium & Exhibition

Email: [email protected] URL: www. punchpowertrain.com In 1985 Patrick Debal graduated as Master of Science in Mechanical Engineering at the University of Leuven, Belgium. He held several positions in research and development before joining Punch Powertrain 2006. At Punch Powertrain Patrick and his team develop a next generation, highly performing hybrid powertrain. In 2009 the first hybrid powertrain from Punch Powertrain was be demonstrated.

ir. Steven Bervoets Punch Powertrain Schurhovenveld 4 125, BE3800 SintTruiden, Belgium Tel: +32 11 679 215 Fax: +32 11 679 230

Email: [email protected] URL: www. punchpowertrain.com Steven Bervoets graduated in 2008 as Master of Science in Electro Technical and Mechanical Engineering at the University of Leuven, Belgium. In September 2008, he joined the Controls Group of Punch Powertrain to develop and test the hybrid control system.

Suggest Documents