摘要
提出了一种带有动态自适应惯性权重和随机变异策略的粒子群优化算法.在每次迭代时,算法可根据粒子的适应度变化动态改变惯性权重,从而使算法具有动态自适应性。当用早熟判断机制判断算法陷入早熟收敛时,采用随机变异策略使其跳出局部最优。将改进的算法应用于GM(1,1,λ)模型的求解,具体实例表明改进的粒子群优化算法能够显著提高GM(11,λ)模型的精度。
This paper proposes a new adaptive particle optimization algorithm with dynamic adaptive inertia weight and random mutation.During the running time,the inertia weight is determined by the particle＇s fitness that makes the algorithm become dynamical and adaptive.By judging the local convergence,when the algorithm gets into the local convergence,it can start random mutation.The improved algorithm is applied to solve the gray model GM（1,1,λ）,experiments on concrete examples show that the modified algorithm has strongly improved the calculation accuracy of gray model GM（1,1,λ）.
出处
《计算机工程与应用》
CSCD
北大核心
2010年第32期44-47,共4页
Computer Engineering and Applications
基金
霍英东教育基金(No.104009)
关键词
GM(1
1
λ)模型
粒子群优化
动态自适应惯性权重
随机变异算子
gray model GM（1,1,λ）； particle swarm optimization； dynamic adaptive inertia weight； random mutation operator；
