一种改进的微粒群算法

Modified particle swarm optimization algorithm

通过在微粒群算法中引入排雷策略的思想,对微粒群优化算法进行改进,使微粒群算法能摆脱局部极值点的束缚;另外通过在算法的迭代过程中加入旋转方向法,加快算法的收敛速度,从而形成一种新的改进粒子群算法。通过对三个典型函数进行优化计算,并与其他文献的改进微粒群算法的优化结果进行比较,表明基于排雷策略的改进算法很好地解决了粒子群优化算法早收敛、难以跳出局部极值点和收敛较慢的问题。

To deal with the problem of premature convergence and slow search speed, this paper proposed a new particle swarm optimization （PSO）. The new method was based on clearing of mines, which was guaranteed to converge to the global optimization solution with probability one. In addition, combined the new method with rotating direction method, which was beneficial for the convergence speed. Through the calculation of three typical function, and made comparison with other improvement particle swarm optimization algorithm, they show that the new method which solves the problem of premature convergence and slow search speed.