An optimization Study for a Switched Reluctance Motor using Magnetic Equivalent Circuit and Space Mapping Techniques

Abstract

Finite Element Method (FEM) provides high accuracy in the design and analysis of electric machines. However, it requires high computational cost and considerable amount of time. Magnetic Equivalent Circuit (MEC) technique is a viable alternative to the FEM as it is faster and has less computational burden at the expense of lower accuracy. In this paper, a magnetic equivalent circuit (MEC) of an 8/6 switched reluctance motor (SRM) is developed. The MEC model is then utilized in a space mapping optimization loop. The space mapping technique compensates the calculation error of the optimized MEC model compared to a Finite Element (FE) model. This approach reduces the computational time as it limits the number of FE simulations. Two different optimization problems are considered. First, the MEC model is optimized to maximize the static torque profile of the considered SRM. Secondly, the model is optimized to achieve a specific torque. The stator and rotor pole arc angles are taken as the design parameters.