GA-Based autonomous electromagnetic design and experimental verification of two-phase switched reluctance compressor drive

This paper presents an approach to genetic algorithm (GA)-based computer aided autonomous electromagnetic design of 2-phase switched reluctance motor drives. The proposed drive is designed for compressor drives of low-priced refrigerators as an alternative to existing brushless DC motor drives with rare-earth magnets. In the proposed design approach, three GA loops work to optimize the lamination design in order to meet the requirements for the target application while simultaneously fine-tuning the control parameters. To achieve the design optimization within a realistic timescale, the repeated-calculation required to obtain fitness evaluation does not use finite element analysis (FEA), but consists of geometric flux tube-based non-linear magnetic analysis and a dynamic simulator based on an analytical expression of the magnetizing curves obtained from the non-linear magnetic analysis. The validity of the proposed design approach is verified through several comparative studies on the design results with FEA and the experimental studies using a 250 W 2-phase 8/6 prototype which is designed by the proposed optimizing GA approach.