ABSTRACT:
This research has investigated automotive electrical systems and discussed the need for and requirements of electromechanical actuation and propulsion in modern automotive vehicles. The low cost, easy assembly, and low rotor inertia of the switched reluctance motor (SRM) make it well suited for many automotive applications. One of the drawbacks of SRMs for the automotive industry is the time to market due to the SRMs nonlinear magnetic operation. This research sought to provide a “first pass?design tool for SRM designers. A D2L sizing equation was developed for SRMs and was programmed in a MATLAB script. Beyond sizing, the MATLAB script also included design guidelines which provide the designer with a complete set of dimensions to model the machine in a more accurate but much more time intensive finite element analysis (FEA) program. The MATLAB script also gives the designer the ability to modify the calculated dimensions and view their effect on output torque.
In developing the sizing program, a large portion of this research was spent investigating maximum and minimum inductance calculations for the SRM. The inductance ratio constant is an important element in the sizing program and cannot be easily determined without some analysis. Three methods for calculating the unaligned inductance were investigated. These methods were implemented and their results compared to experimental and FEA results. None were shown to produce results closer than 18% for the motor used in this research. Three methods were implemented to calculate the aligned inductance. Two of these were relatively complex and ultimately never produced accurate results. In an effort to understand the problem a third, simple model was created and shown to calculate the aligned inductance with a 4-10% error.
The completed sizing program was verified using speed, torque, current, and efficiency data collected from the studied motor, a 42 V motor capable of providing 2.5 Nm at 4000 rpm. Over the full range of torque, the error between both calculated and measured efficiency and between the calculated speed and measured speed varied from -33% to 20%. This may be an acceptable starting point in some. The sizing program was then used to generate a complete geometry for 3 given points on the SRMs speed torque curve. These results showed considerable variation from the designed machine at higher torques and the reason for this remains to be determined.
If your company is a member of the Mechatronic Laboratory, please send the request to receive a copy of any technical report. If you are not a member please send a request Ali Keyhani, Department of Electrical Engineering, Mechatronics Program at the following address: Ali Keyhani, Ohio State University, Electrical Engineering Department, Mechatronics Systems Laboratory, 2015 Neil Ave., 205 Dereese Lab., Columbus, OH 43210.