摘要
万有引力搜索算法(gravitational search algorithm,GSA)相比于传统的优化算法具有收敛速度快、开拓性能强等特点,但GSA易陷入早熟收敛和局部最优,搜索能力较弱.为此,提出一种基于改进的Tent混沌万有引力搜索算法(gravitational search algorithm based on improved tent chaos,ITC-GSA).首先,改进Tent混沌映射来初始化种群,利用Tent混沌序列随机性、遍历性和规律性的特性使得初始种群随机性和遍历性在可行域内,具有加强算法的全局搜索能力;其次,引入引力常数G的动态调整策略提高算法的收敛速度和收敛精度;再次,设计成熟度指标判断种群成熟度,并使用Tent混沌搜索有效抑制算法早熟收敛,帮助种群跳出局部最优;最后,对10个基准函数进行仿真实验,结果表明所提算法能够有效克服GSA易陷入早熟收敛和局部最优的缺点,提高算法的收敛速度和寻优精度.
The gravitational search algorithm(GSA)has the characteristics of faster convergence speed and stronger exploitation performance than the traditional optimization algorithm,but the GSA is vulnerable to premature convergence and local optimum,and its search ability is weak.Therefore,this paper proposes the gravitational search algorithm based on improved Tent chaos(ITC-GSA).Firstly,the Tent chaotic map is improved to initialize the population.Using the characteristics of randomness,ergodicity and regularity of Tent chaotic sequence,the initial population randomness and ergodicity are within the feasible domain,and the global search ability of the algorithm is enhanced.Then,the dynamic adjustment strategy of the gravity constant G is introduced to improve the convergence speed and convergence accuracy of the algorithm.Moreover,maturity indicators are designed to determine the maturity of the population,and Tent chaos search is used to effectively suppress the premature convergence of the algorithm for helping the population jump out of the local optimum.Finally,through simulations of 10 benchmark functions,experiments show that the proposed algorithm can effectively overcome the shortcomings of the GSA’s vulnerability to premature convergence and local optimization,and improve the algorithm’s convergence speed and optimization accuracy.
作者
张娜
赵泽丹
包晓安
钱俊彦
吴彪
ZHANG Na;ZHAO Ze-dan;BAO Xiao-an;QIAN Jun-yan;WU Biao(School of Informatics and Electronics,Zhejiang Sci-Tech University,Hangzhou 310018,China;Guangxi Key Laboratory of Trusted Software,Guilin University of Electronic Technology,Guilin 541004,China)
出处
《控制与决策》
EI
CSCD
北大核心
2020年第4期893-900,共8页
Control and Decision
基金
国家自然科学基金项目(61502430,61562015)
广西自然科学重点基金项目(2015GXNSFDA139038)
浙江理工大学521人才培养计划项目.