基于混沌序列的多峰函数微粒群寻优算法的目标就是找到多峰函数的所有局部优化峰值。在分析微粒群优化

算法中各个参数对微粒运动影响的基础上，对微粒群算法进行改造，让微粒运动从初始位置沿优化函数曲线向优化峰值

方向爬行．直至找到所在区域的局部优化峰值；要想求得尽可能多的局部优化峰值，就要求微粒群中微粒的初始位置分

布具有随机性和遍历性。为此采用混沌序列设置微粒初始位置；为使每一个局部最优值点都可能有微粒群中的微粒经过，

采用变步长的迭代计算；为防止优化函数曲线的某些局部峰附近没有微粒分布，从而漏掉该局部峰值，对计算进行重复，

直至两轮求得的优化函数的局部峰值之差小于给定阈值。仿真结果表明，该算法具有很好的局部寻优特性，计算过程简

捷，寻优效果良好，可有效地应用于多峰函数的局部寻优并求取全局最优值-Based on chaotic sequence of multimodal function particle swarm optimization algorithm goal is to find a multimodal function of all local optimization peak. On the analysis of particle swarm optimization

Algorithm of various parameters on the particle motion influence on the basis of the particle swarm algorithm transform, let particle motion from the initial position along the optimization function curve to optimization peak

Crawling direction. To find the area of the local optimization peak　If you want to get as many local optimization peak, requires the particle swarm the particles in the initial position points

Cloth has the nature of randomness and ergodicity. Therefore the chaotic sequence set particle initial position　In order to make every local optimal value point may have particle swarm particles after,

The variable step long iterative calculation　To prevent optimization function curve of some local peak no nearby particle distribution, thus miss this local peak, the c