一种改进的连续域蚁群算法  

An improved Ant Colony Optimization for continuous domains

作　　者：李抒玹 付远望 徐光辉[2] LI Shuxuan;FU Yuanwang;XU Guanghui(Hubei University of Technology Detroit Green Technology Institute,Wuhan 430068,China;Hubei University of Technology School of Electrical and Electronic Engineering,Wuhan 430068,China)

机构地区：[1]湖北工业大学底特律绿色工业学院,武汉430068 [2]湖北工业大学电气与电子工程学院,武汉430068

出　　处：《中南民族大学学报（自然科学版）》2024年第6期816-826,共11页Journal of South-Central Minzu University(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61603127)。

摘　　要：针对连续域蚁群算法(ACOR)陷入局部最优、收敛精度较差等缺陷,提出了一种改进连续域蚁群算法.首先,通过改进ACOR算法选择指导解的方式增强种群多样性;其次,利用惯性权重对均值更新公式进行了改进,并对ACOR算法中ζ参数进行线性递减处理;最后,为了帮助算法在陷入停滞时尽快跳出局部最优,用差分变异策略对算法的搜索进行监控,搜索陷入停滞时产生新解,并通过筛选出更好的解跳出局部最优.设计了两种维度环境的实验,其中包括16个可变维度函数和4个固定维度函数,并展示了改进算法与其他几种算法的寻优结果.结果表明,所提出的改进算法在搜索的过程中有效地缓解了上述缺陷.An improved Ant Colony Optimization for continuous domains(ACOR)was proposed to overcome the shortcomings of ACOR,such as falling into local optimization and poor convergence accuracy.Firstly,the diversity of the population was enhanced by improving the way the guidance solution selected by ACOR.Secondly,the inertia weight was used to improve the mean value updating formula,and the parameterζ,which was a parameter in ACOR algorithm,was linearly decreased.Finally,in order to help the algorithm jump out of the local optimum as soon as possible when it was stuck,a differential mutation strategy was used to monitor the algorithm's search,new solutions were generated when it was stuck,and better solutions were filtered out to jump out of the local optimum.The experiments were designed in two dimensional environments,including 16 variable dimensional functions and 4 fixed dimensional functions,and the optimization results of the improved algorithm and other algorithms were shown.The results indicated that the above defects in the search process were effectively alleviated by the improved algorithm.

关 键 词：连续域蚁群算法 收敛精度 指导解选择 惯性权重 差分变异 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TP181[自动化与计算机技术—计算机科学与技术]