第5章应用matlab实现遗传算法优化二自由度pid控制 - 进入正式系统.doc

摘 要 遗传算法是密歇根大学Holland教授借鉴生物进化中的“生存竞争”和“优胜劣汰”现象提出的有效的全局优化算法。它将遗传操作应用于一群对搜索空间编码的染色体中，在每一代，遗传算法同时作用于整个搜索空间的不同区域，通过优胜劣汰，去掉解空间中期望值较低的部分，保留高期望值部分，从而能以较大的概率找到最优解。 遗传算法（GA）是模仿自然选择、物种进化和群体遗传学而建立的一种随机搜索技术,其主要特点是群体搜索策略和群体中个体之间的信息交换，搜索不依赖于梯度信息。由于遗传算法不受搜索空间的限制性假设的约束，不要求解空间有连续性、可导等性质，以及其固有的并行性。迄今为止，己经在工程和研究的诸多领域得到了广泛的应用。 本文首先介绍了遗传算法的发展、特点、应用、实现过程及其应用。介绍了PID控制技术的原理和应用，最后本文应用MATLAB软件实现了遗传算法的基本操作（包括复制，交叉，变异），并且实现了遗传算法对函数的优化和对二自由度PID控制器参数的整定，并和传统方法整定PID的效果进行了比较。结果表明,遗传算法比传统方法优化PID参数的效果更好。 遗传算法；二自由度PID控制器；MATLAB；参数整定Genetic algorithms (GA) is a random search technology which imitates natural selection, species evolution and population genetics. They search from and maintain a population ,use Payoff information among themselves, all of which are the main advantage of the algorithms. GA don’t require the solution space to be continuous and derivable. It’s few limitation on the presumption of the solution space and built-in parallelism make it a useful method in many area. So far, it has been widely used in engineering and research in many fields. Genetic algorithms (GA)is a sort of global, efficient optimization algorithms which was put forward by professor Holland in Michigan University inspired by the phenomena of survival competition and natural selection. In the application of GAs, genetic operators operate on a group of chromosome which is coded in search space, In each generation, Gas operates on the entire search space simultaneously and preserves the fitter individuals imitating natural selection, resulting in the fact that the optima can be found by high probability. This paper introduces the development, features, applications and the implementations of genetic algorithm, and its applications. This paper also presents the theory and applications of PID control technology. Finally, this paper uses MATLAB software to realize the basic operations of genetic algorithm, including reproduction, crossover and mutation, and employs genetic algorithm to optimize function and the param