基于增量式PID的改进粒子群算法

基于粒子群优化(PSO)算法的简单模型和增量式PID控制原理,引入PID增量算子和4个新随机因子,对标准粒子群优化(SPSO)算法进行了扩展.扩展粒子群算法(EPSO)提升了粒子自身认知能力和社会认知能力,增加了粒子共享的信息量,粒子在运动过程中更加智能化.4个新随机因子的引入,提高了种群的多样性,一定程度上克服了PSO容易陷入局部最优的缺陷,提高了PSO算法全局搜索能力.对6个常用目标函数进行优化仿真,结果表明EPSO算法较SPSO算法收敛速度显著加快,且不易陷入局部极值点.SPSO算法是EPSO算法的一种特殊情形;EPSO算法作为SPSO的扩展,可应用于所有SPSO求解的优化问题.

Modified Particle Swarm Optimization Based on Increment PID

Based on the analysis of the standard Particle Swarm Optimization (SPSO) model, a new extended PSO(EPSO) algorithm was proposed. By introducing Increment PID operator and four more random factors in EPSO, the new algorithm improves the individual and social cognition ability of particles, increases shared information and makes the motion of particles more intelligentized. Furthermore, the population variety has been improved with four more random factors, and then the global optimization has been greatly improved. Six common used test functions have been optimized by both improved algorithm and SPSO. Simulation results have verified the superior performances of the EPSO algorithm. As an expansion of SPSO algorithm, the EPSO is universal and can be used to solve all optimization problems in power systems with SPSO.