神经网络PID控制器的设计与仿真.docVIP

*******大学 毕 业 设 计(论文) 题 目 神经网络PID控制器的设计与仿真 院 系 专业班级 学生姓名 指导教师 二○○八年六月神经网络PID控制器的设计与仿真 摘要 PID控制技术是一种应用很普遍的控制技术,目前在很多方面都有广泛的应用. 在工业控制中，PID控制是工业控制中最常用的方法。这是因为PID控制器结构简单、实现简单，控制效果良好，已得到广泛应用。据统计，在目前的控制系统中，PID控制占了绝大多数。但是，他具有一定的局限性：当控制对象不同时，控制器的参数难以自动调整以适应外界环境的变化。为了使控制器具有较好的自适应性，实现控制器参数的自动调整，可以采用神经网络控制的方法。利用人工神经网络的自学习这一特性，并结合传统的PID控制理论，构造神经网络PID控制器，实现控制器参数的自动调整。讨论了基于神经网络的PID控制，利用神经网络的自学习能力进行在线参数整定，并利用Matlab软件进行仿真。PID控制　神经网络　Matlab The design and simulation of the neural network PID controller ABSTRACT PID control technology is a very common control technology in many aspects of a wide range of applications. In industrial control, PID control is the most commonly used in industrial control methods. This is because the PID controller simple structure, to achieve a simple, effective control, has been widely used. According to statistics, PID control is the vast majority in the present control system. However, he has certain limitations: When the control object is not at the same time, the controller parameters to automatically adjust to the changes in the external environment. In order to make the controller has good adaptability, and Controller Parameters of automatic adjustments can be used neural network control method. Using artificial neural network learning oneself, combined with the traditional PID control theory Structure of neural networks PID controller, and implementate the automatic adjustment of controller parameters. The thesis discussed according to the neural network PID controller，and control，make use of the neural network from the study ability to proceed its function to on-line parameter amend，and make use ofthe Matlab software proceeds to imitatereally．By the simulation can see that it is to achieve self-learning, adaptability, and other characteristics ，network convergence speed, can have a good control on non-linear object ,and tracking of system performance very good. Its parameters need to know the object of the specific parameters