Multi-objective optimization design method for passive filter in multiisland particle swarm optimization (PSO)-based mixed type filter

Abstract

The invention provides a multi-objective optimization design method for a passive filter in a multiisland particle swarm optimization (PSO)-based mixed type filter. The method comprises the steps of: determining multiisland structure of the initial swarm in the PSO; improving the PSO to particle swarms on islands; determining the best position found so far in particles of the islands, updating the fitness function corresponding to the best position; and determining the size of combination interval and the combination rate among islands, selecting particles with optimal fitness among the islands to be combined into new islands, sequencing the fitness of the particles, and improving the PSO by new initial point; and judging the termination conditions. Proved by simulation and experiment results, the method has stronger global search capability, can avoid locally optimal solution as much as possible, and achieve the purposes of restraining premature convergence and improving the PSO convergence speed.

Description

technical field [0001] The invention relates to a multi-objective optimal design method for a passive power filter, in particular to a multi-objective optimal design method for a passive filter in a hybrid filter based on a multi-island particle swarm algorithm. Background technique [0002] The hybrid active power filter (HAPF for short) composed of passive filter banks and active filters in parallel has many advantages such as low cost and good filtering effect. important role played. In the hybrid power filter, the passive filter undertakes the main tasks of suppressing harmonics and compensating reactive power, while the active filter is only used to improve the filtering effect of the passive filter and play a role of filling in gaps. Among them, the optimal design of the passive filter in the hybrid system largely determines whether the hybrid power filter can achieve the expected harmonic suppression and reactive power compensation effects according to its working pr...

AI Technical Summary

Problems solved by technology

In the third issue of "Chinese Journal of Electrical Engineering" in 2002, the article "Multi-objective Optimal Design of Passive Filters" proposed a genetic algorithm with a chaotic operator, which overcomes the premature convergence of the algorithm by using the chaotic operator of the algorithm. The disadvantage is that the algorithm development cycle is longer and the operation is more complicated

The article "Optimization Design of Passive Filter Based on Interactive Multi-objective Genetic Algorithm" in the 6th issue of "Acta Electrotechnical Society" in 2003 proposed an interactive genetic algorithm, which guided the genetic search through inaccurate preference information, so that the algorithm population moved toward The best comprehensive performance of multiple objectives evolves, increasing the diversity of understanding, but the disadvantage is that the algorithm is more complicated

The disadvantage is that the algorithm program takes up a lot of system memory space, and it is difficult to implement

Images