混合策略改进的风驱动优化算法

Wind Driven Optimization Algorithm Improved by Hybrid Strategy

为解决风驱动优化算法存在的易陷入局部极值及收敛性差等问题,提出一种混合策略改进的风驱动优化算法。首先,使用Tent混沌映射初始化种群,增加初始个体的多样性;其次,引入柯西变异策略,扩大算法搜索范围,增强算法搜索能力并加速算法收敛;然后,利用反向学习策略生成新的全局最优解,提高算法逃离局部极值能力;最后,针对6个基准测试函数进行仿真实验,结果表明,所提算法收敛速度和精度均优于其他算法。

In order to solve the problems of wind driven optimization algorithm,such as easy to fall into local extremum and poor convergence,an improved wind driven optimization algorithm combining Tent mapping,Cauchy mutation and opposition-based learning strategy is proposed.Firstly,the algorithm used the Tent mapping to initialize the population,which increased the diversity of the initial individuals.Secondly,the Cauchy mutation strategy is introduced to expand the search range of the algorithm,enhance its search ability and accelerate its convergence.Then,opposition-based learning strategy is used to generate a new global optimal solution,which improves the ability of the algorithm to escape from local extremum.Finally,simulation experiments are carried out on six benchmark test functions.The results show that the proposed algorithm has better convergence speed and accuracy than other algorithms.