Optimal design of switched reluctance motor using genetic algorithm

Abstract EN

Switched reluctance motor (SRM) has increased interest in both research and industry due to its unique characteristics.

the simple structure without windings or permanent magnets on the rotor makes the motor robust, reliable, and most importantly a low manufacturing cost.

SRM also provides high starting torque and high efficiency over a wide range of speeds which is strongly desired in electric vehicles application.

however, these advantages of switched reluctance motors come with some challenges.

torque ripples, low power density, and temperature rise are common problems in SRMs.

in this paper, multi-objective optimization of SRM design is achieved to obtain most of the SRM desired characteristics with minimization of the machine's common drawbacks.

the optimization process considers twelve variables and five objective functions.

objective functions include average torque, efficiency, iron weight, torque ripples, and maximum temperature rise.

the electromagnetic analysis of each candidate is performed by the finite elements method (FEA).

the performance indices of SRM are calculated based on FEA analysis results by calculations that achieve both accuracy and speed.

the optimization is achieved by the multi-objective genetic algorithm technique (MOGA) in which the multi-objective functions are converted to a single objective function.

the efficiency map, torque profile, and dynamic simulation of the motor is provided as a verification of the optimal design.

this paper mainly studies the design and optimization of SRM.

the design and optimization process aim to fulfill the general requirements of electric vehicle application.