一种动态调整惯性权值的粒子群算法

粒子群算法针对速度变量的调节不够精确,算法在迭代过程中容易陷入局部最优,函数目标值的精度比较低。为了得到更加精准的目标值,提出一种改进的粒子群算法,对粒子群算法的惯性权值参数进行动态调整。算法将惯性权值参数设置为由粒子位置、个体最优位置和全局最优位置影响的可变参数组,通过各个位置之间的距离来控制参数的改变。该改进粒子群算法针对每一个粒子的每一维度都设计其对应的惯性权值参数。该改进算法经过和其它算法进行比较测试,结果显示改进的算法得到的解值更加精准。

A Particle Swarm Optimization Algorithm with Dynamically Adjusting Inertia Weight

Particle swarm optimization(PSO)is not accurate enough to adjust the variable of speed,the result of PSO is easy to fall into local optimum,and the accuracy of the fitness value is relatively low.In order to get more accurate fitness value,this paper improves PSO by changing the inertia weight parameters dynamically.The algorithm sets the inertia weight as parameters related to particle position,individual optimal position and global optimal position.The change of parameters is controlled by the distance between each position,and every dimension of every particle has its own inertia weight parameter.Compared with other algorithms,the results show that the improved algorithm is more accurate.