Optimal Design for Rectifier Transformer Using Improved PSO Algorithm

An improved partial swarm optimization algorithm (IPSO) that can overcome the disadvantages of trapping in local optimum in the standard PSO algorithm is proposed for rectifier transformer optimal design. In the calculation process of IPSO, the max iteration is divided into a number of sections by control factor. At the beginning of each section, the single particle inertia weight is determined by the comparing result between its own fitness function value and the average fitness function value of the particle swarm. Therefore, the particles with good fitness function value will trend to make local search, and the ones with poor fitness function value will do rough detection in the completely feasible region to discover new better solution. Therefore, the IPSO has a good effort to adjust the dynamic balance between swarm´s convergence and individual´s diversity. A systematic optimal design method of rectifier transformer was proposed based on the IPSO. The feasibility and validity of the proposed method are proved by the contrast between handwork results and IPSO results.