论文部分内容阅读
为了克服粒子群优化容易陷入局部极小的缺陷,利用粒子速度不依赖于其与最优粒子之间距离的大小,而仅依赖其方向信息的特点,采用自适应策略弹性地修正粒子速度的幅值.同时,充分利用混沌运动的遍历性、随机性及对初值的敏感性等特性,提出一种基于混沌的弹性粒子群优化(CRPSO)算法,并将其成功用于典型多极点函数优化.仿真结果表明,该算法增强了摆脱局部极值点的能力,提高了收敛速度和精度.
In order to overcome the particle swarm optimization easy to fall into the local minimum defects, the use of particle velocity does not depend on its distance from the optimal particle size, but only depends on the characteristics of its directional information, using adaptive strategy to amend the elastic particle velocity In the meantime, taking full advantage of chaotic ergodicity, randomness and sensitivity to initial value, a chaos-based elastic particle swarm optimization (CRPSO) algorithm is proposed and applied to the optimization of typical multipole function The simulation results show that this algorithm enhances the ability to get rid of local extremum points and improves the convergence speed and precision.