A Multi-stage Evolutionary Algorithm for Solving Complex Function Optimization Problems

Abstract

Based on the analysis of defects of traditional evolutionary algorithms in solving global optimization of non-linear or multi-modal function, a novel evolutionary algorithm called Multi-Stage Evolutionary Algorithm (MSEA) is proposed. MSEA has many new features. It develops some new operators such as multi-parent crossover operator with elite-preservation, dynamical mutation operator, space contraction operator, etc; It introduces a new multi-stage algorithm framework. The simulation results on some typical test problems show that MSEA proposed in this paper is better than existing evolutionary algorithm in the accuracy of solutions.