多策略融合的改进风驱动优化算法  

Improved Wind Driven Optimization Algorithm Based on Multi-Strategy Fusion

作　　者：陈伟[1] 吴宣够 CHEN Wei;WU Xuangou(Department of Computer Information,Suzhou Vocational and Technical College,Suzhou Anhui 234909,China;School of Computer Science and Technology,Anhui University of Technology,Maanshan Anhui 243032,China)

机构地区：[1]宿州职业技术学院计算机信息系,安徽宿州234909 [2]安徽工业大学计算机科学与技术学院,安徽马鞍山243032

出　　处：《重庆科技学院学报（自然科学版）》2022年第5期44-52,共9页Journal of Chongqing University of Science and Technology：Natural Sciences Edition

基　　金：安徽省教育厅自然科学基金重点项目“基于Spark的电商网站用户行为分析预测系统研究”(KJ2019A1058);安徽省高校优秀青年骨干国内访学者研修项目(GXGNFX2020152)。

摘　　要：针对风驱动优化(WDO)算法易陷入局部最优值的问题,将Tent混沌映射、参数自适应调整及小波变异等策略融入WDO算法中,提出了一种多策略融合的改进风驱动优化(MFWDO)算法。首先,利用Tent混沌映射完成种群初始化,以增加种群个体的多样性;其次,引入动态因子对WDO算法中的固有参数进行自适应调整,以加快算法的收敛速度;最后,对每代最优粒子进行小波变异,以更新种群中较差个体的位置,使算法跳出局部最优,从而提高算法的全局搜索能力。由9个标准测试函数的仿真实验结果可知,MFWDO算法的求解精度和收敛速度较其他算法明显提升。通过对压力容器、焊接梁等工程优化问题进行实验,进一步验证了MFWDO算法的有效性和优越性。Aiming at the problem that the wind driven optimization(WDO)algorithm is easy to fall into the local optima,a multi-strategy fusion wind driven optimization algorithm(MFWDO)integrating with tent chaotic mapping,parameter adaptive adjustment and wavelet mutation is proposed.Firstly,tent chaotic mapping is used to initialize population,which increases the diversity of initial population.Secondly,the dynamic factor is introduced to adjust the parameters of WDO,which improves the convergence speed of the algorithm.Then,wavelet mutation is performed to mutate the optimal particle of each generation to update the position of poor individuals in the population,which helps the algorithm to jump out of the local optima and improves its global search ability.The simulation results on 9 standard functions show that the solution accuracy and convergence speed of MFWDO are significantly improved compared with other algorithms.The effectiveness and superiority of MFWDO are further verified by optimizing the two practical engineering problems of pressure vessel design and welded beam design.

关 键 词：风驱动优化算法 Tent混沌映射 小波变异 自适应调整 工程优化 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]