Particle Swarm Optimization Techniques Applied to the Design of a Digital PMSM Servo Drive

摘要

This paper presents a novel design approach for a permanent-magnet synchronous machine in a HEV (hybrid-electric vehicle) by applying IPSO (Improved Particle Swarm Optimization) to optimize the fuzzy input and output scaling factors k{sub}a ＞ k{sub}b ＞ k{sub}u of FLC (Fuzzy Logical Controller) in different operating regions, while keeping the fuzzy rules and membership functions fixed. The proposed scheme is derived from the detailed analysis of the general dynamics of the drive system, its effectiveness has been verified by simulation with MATLAB. The system is tested using a step change signal of load. The simulation results show that the use of IPSO as an optimization algorithm makes the drive robust, with faster dynamic response, higher accuracy and intensitive to load variation.