Differential Evolution with Grey-based Hybrid Trial Vector Generation Strategies

Ming-Feng Yeh ； Min-Shyang Leu ； Ti-Hung Chen ； Po-Jung Huang

《Journal of Grey System》 16卷4期 (2013/12) Pp. 251-258

並列摘要

This study attempts to propose a novel grey-based trial vector generation strategies to improve the search performance of differential evolution (DE) algorithm. The proposed strategies involve two main components. One is our previous work, differential evolution with grey-based adaptive mutation factor (DE_GAMF), and the other is the hybrid crossover strategy. The hybrid crossover strategy consists of a pool of distinct crossover strategies and randomly selects one to generate the trial vector for the associated target vector. DE with grey-based hybrid trial vector generation strategies are also applied to solve the optimization problems of eight benchmark functions for illustration. Simulation results show that the proposed approach could perform well search performance on most of the test functions.

並列關鍵字

Control parameters ； Crossover strategy ； Differential evolution ； Grey relational analysis ； Mutation strategy