摘要
风驱动优化算法是一种新兴的基于群体的迭代启发式全局优化算法。针对风驱动优化算法易陷入局部最优值的问题,实现了5种带有不同变异策略的风驱动优化算法,这些变异策略分别是小波变异策略、混沌变异策略、非均匀变异策略、高斯变异策略以及柯西变异策略。应用不同变异策略的风驱动优化算法对不同维度的经典测试函数进行了仿真实验,并与粒子群优化算法进行了比较。实验结果表明,小波变异风驱动优化算法具有较强的开发能力,可有效跳出局部最优,其寻优速率、收敛精度及算法稳定性均优于粒子群优化算法、风驱动优化算法和其他改进算法。
The wind driven optimization（WDO）algorithm is a population-based iterative heuristic global optimization algorithm.However,in order to deal with the problem that WDO algorithm is easily trapped into local optima,we introduced five WDO algorithms based on different mutation strategies.They are wavelet mutation strategy,chaotic mutation strategy,non-uniform mutation strategy,Gaussian mutation strategy and Cauchy mutation strategy.Different WDO mutation strategies were used to implement simulation experiments for several typical test functions and compared with particle swarm optimization（PSO）algorithm.Experiments show that the WDO with wavelet mutation（WDOWM）algorithm has a strong developing ability,which has capability to jump out of the local optima.The WDOWM algorithm is superior to the PSO algorithm,WDO algorithm and other improved WDO algorithms in terms of convergence rate,convergence accuracy and stability.
出处
《计算机科学》
CSCD
北大核心
2016年第1期275-281,305,共8页
Computer Science
基金
国家自然科学基金项目(61401179)资助
关键词
风驱动优化算法
变异
全局优化
Wind driven optimization algorithm
Mutation
Global optimization
