A Comparative Study of Shape Optimization of SRM using Genetic Algorithm and Simulated Annealing

This paper describes the Shape Optimization of Switched Reluctance Machine (SRM) using Genetic Algorithm (GA) and Simulated Annealing (SA). To achieve the required performance within a specified space envelope, the physical dimensions of the Switched Reluctance Machine like Stator pole arc, Rotor pole arc, Rotor diameter and Stack length were optimized using GA. The proposed strategy improves the Power Density of the SRM by 11.7% using GA. Similarly using SA, the power density of the machine is increased by 26.94%. The proposed design, in comparison with standard design procedures, highlights improvement in performance with considerable reduction in size. Both the methods GA and SA maximize Flux linkage and Torque per unit rotor volume of the SRM. Even in very high power applications such as Aerospace applications, it is possible to achieve similar optimization using the proposed strategy. The simulation results obtained for a 4 phase, 8/6 pole, lkW, 100V, 25A, 1500 rpm SRM signify the usefulness and effectiveness of the proposed strategy.