论文部分内容阅读
为避免早熟收敛和提升粒子在高维空间的搜索能力,文章提出了一种“自我”感知的高维混沌群体智能算法。首先,采用pBest和gBest混沌双扰动来增强粒子的搜索能力;其次,提出一种“自我”感知策略来帮助种群避免早熟收敛;最后,将三种不同微粒群优化(Particle Swarm Optimization,PSO)算法在旅行推销员问题(Traveling Salesman Problem,TSP)上进行了对比实验。实验结果显示“自我”感知的高维混沌群体智能算法简单、有效可行,值得推荐。
In order to avoid premature convergence and improve the search ability of particles in high-dimensional space, a high-dimensional chaotic group intelligence algorithm named “self” is proposed. First, the chaotic and two-perturbations of pBest and gBest are used to enhance the search ability of particles. Secondly, a “self-aware” strategy is proposed to help the population avoid premature convergence. Finally, three different Particle Swarm Optimization (PSO) ) Algorithm is compared with Traveling Salesman Problem (TSP). The experimental results show that the intelligent algorithm of high-dimensional chaotic population with “self” perception is simple, effective and feasible, which is worth recommending.