论文部分内容阅读
针对微粒群算法(particle swarm optimization)收敛速度慢和早熟收敛的问题,提出一种基于二级搜索(Two steps search)和高斯学习(Gauss learning)相结合的粒子群优化算法(TGPSO).该算法借鉴人工蜂群算法能有效地进行局部搜索和全局搜索,并能在陷入局部极值时跳出局部极值的特点,从两方面对微粒群算法进行改进:通过二级搜索,强化较优粒子的局部搜索能力,可加快收敛速度;应用高斯学习的自适应逃逸能力,可有效地逃离局部最优点.在典型测试函数集上的仿真实验结果表明本文算法有较好的寻优性能并能快速地找到最优解.
To solve the problem of slow convergence and premature convergence of particle swarm optimization, a particle swarm optimization (TGPSO) algorithm based on two steps search and Gauss learning is proposed. By using artificial bee swarm algorithm, local search and global search can be effectively carried out, and the local extremum can be popped out when it falls into local extremum. The improved particle swarm optimization is improved from two aspects: Local search ability can speed up the convergence rate and Gaussian learning adaptive escape ability can effectively escape from the local optimum point.The simulation results on a typical test function set show that the proposed algorithm has better performance and can quickly Find the optimal solution.