An Adaptive Learning Co-Evolutionary Variational Particle Swarm Optimization Algorithm for Parameter Identification of PMSWG.

Targeting the problems of traditional particle swarm algorithm easily falling into local optimum and low recognition accuracy, an adaptive learning co-evolutionary variational particle swarm optimization algorithm (ALCEVPSO) is proposed in this paper to identify the parameters of permanent magnet synchronous wind generator (PMSWG). At first, an adaptive learning strategy is adopted for the inertia weights of the PSO, and the global optimization seeking ability of the PSO is improved. After that, multiple swarm co-evolution strategies are introduced to share the best positions within sub-populations, and by this method, the algorithm’s falling into local optimality is avoided. Finally, Cauchy Gaussian mixed variants are introduced, and the population diversity is enriched. The proposed method has the advantages of strong optimization ability and high search accuracy compared with the traditional particle swarm algorithm, which is shown by simulated and experimental results. By this method, the motor parameters of the permanent magnet synchronous motor can be accurately identified. [ABSTRACT FROM AUTHOR]