论文部分内容阅读
在两种微粒群算法分析的基础上,针对算法存在局部最优和后期振荡的现象,提出一种改进自适应微粒群算法.新算法引入概率突跳因子改变了原算法中微粒的速度更新公式,引入模拟退火接受准则抑制了概率突跳的不可控制性.典型函数寻优结果表明,新算法能很快地收敛到全局最优解,大幅度降低了达到最优值所需要的迭代数,同时提高了算法的收敛率和收敛精度,在跳出局部搜索的能力上远优于标准微粒群算法和自适应微粒群算法,稳定性好.
Based on the analysis of the two particle swarm optimization algorithms, aiming at the phenomenon of local optimum and post-oscillation of the algorithm, an improved adaptive particle swarm optimization algorithm is proposed. The new algorithm introduced the probability kurtosis factor to change the particle velocity update formula , The introduction of simulated annealing acceptance criterion suppresses the uncontrollability of probability kurtosis.Experimental results of typical functions show that the new algorithm converges to the global optimal solution quickly and greatly reduces the number of iterations required to reach the optimal value, At the same time, the convergence rate and convergence accuracy of the algorithm are improved, which is much better than the standard PSO and PSO in jumping out of the local search. The stability is good.