基于PSO算法的PID设计.doc

A PID Design Based on the Particle Swarm Optimization algorithm 摘 要 当今社会中，随着科技、知识、工业水平的不断进步，对于各项技术的要求也不断在提高，人们对于先进技术的改进的关注也与时俱增。在工业控制领域，在控制难度、指标要求、复杂程度都在不断提升的前提下，对工业控制技术也提出了新的期盼。以往例如Z-N法等经验法已经不能满足最新的控制要求，而在遗传算法提出的几十年里，计算精度与效率已不容人们忽视。各种基于遗传算法的改进算法频繁出现，但工程前线的工程师们仍在不停探索能够得到普遍适用的算法。粒子群优化算法(PSO)是一种进化计算技术，由Eberhart博士和Kennedy博士提出。源于对鸟群捕食的行为研究，PSO算法假想一群鸟在随机搜索食物。通过粒子模拟鸟个体搜索食物，根据个体的位置、速度、适应度等参变量求得个体最优和全局最优两个极值。是一种已得到验证的有巨大潜力的进化优化算法。通过一种基于PSO算法的PID控制器设计，得出仿真结果验证其优势。 关键词 PID PSO 工业控制 粒子群算法 优化 Abstract In todays society, accompanied with the continuous improvement of science and industry, it becomes more important for the requirements of technical. Consequently people pay more and more concern on the increase of advanced technology for the past years. In the field of industrial, difficulty in controlling, target and a rising complexity provide a push to the industrial control technology just as the new expectations. Previous experience such as ZN law can no longer meet the latest control requirements, and in the genetic algorithm has been put forward for several decades, the accuracy and efficiency have been not able to satisfy the need. A variety of improved algorithms based on genetic algorithms were brought forward frequently, but none of them could be generally applicable. Particle swarm optimization (PSO) is an evolutionary computation technique, put forward by Dr. Eberhart and Dr. Kennedy. Imitating from the behavior of birds, PSO algorithm suppose a group of birds random search for food, which is the single or several best. In the particle simulation of birds searching for food, according to the individuals location, speed, fitness and other individual variables we can obtain the two polar value—particle optimum and global optimum. The PSO algorithm has been proven to be evolutionary optimization algorithm with a huge potential. This article, experimenting PID controller design based on PSO algorithm, shows its advantages through the simulation results. Keywords PID P