论文部分内容阅读
为了提高群集蜘蛛优化(SSO)算法的性能,提出一种基于动态学习策略的群集蜘蛛优化(DSSO)算法.该算法通过群体协作过程中学习因子的动态选择,平衡算法的搜索能力和勘探能力;采用随机交叉策略和云模型改进协作过程个体更新方式,在维持种群多样性的同时尽量提高收敛速度.基于标准测试函数的仿真实验表明,DSSO算法可有效避免早熟收敛,在收敛速度和收敛精度上较标准SSO算法和其余4种较具代表性的优化算法均有显著提高.
In order to improve the performance of Cluster Spider Optimization (SSO) algorithm, this paper proposes a cluster spider optimization (DSSO) algorithm based on dynamic learning strategy, which dynamically selects the learning factors and balances the search ability and exploration ability of the algorithm in the collaborative process of group collaboration. The random crossover strategy and the cloud model are used to improve the individual updating mode of the collaborative process, and the convergence rate is kept as high as possible while maintaining the population diversity.A simulation experiment based on the standard test function shows that the DSSO algorithm can effectively avoid premature convergence, and the convergence speed and convergence accuracy Compared with the standard SSO algorithm and the remaining four representative optimization algorithms have been significantly improved.