基于PLC的神经网络PID控制器设计.docVIP

中 北 大 学 2013 届 毕 业 设 计 说 明 书 第 PAGE 45页 共 NUMPAGES 45页 基于PLC的神经网络PID控制器设计 摘要：PID控制是控制领域中最为重要的控制方式之一。随着对人工神经元网络研究的深入研究，研究者也开始将神经网络和PID控制结合，以期为改进和克服传统PID控制的性能提供了新的思路。 本论文在引入了神经网络及PID神经元网络的相关概念之后，对PID神经元 网络模型和算法进行了系统的介绍。针对常规PID控制的温室环境控制系统存在参数难以整定、超调量大、调节时间长、抗干扰能力差等问题，利用神经网络对系统性能的自学习优点找到能使系统性能达到最佳的PID控制参数，并将其直接送给常规PID控制器，得到了基于BP神经网络整定的PID控制器，从而有效的解决了常规PID存在的问题。 利用MATLAB软件对BP神经网络PID控制进???仿真，编写了仿真程序，对阶跃响应和给定突变两种情况进行了仿真分析。最后简单介绍了西门子S7-300 PLC ，并通过其开发环境编写了PID控制的梯形图。 关键词：PLC; BP神经网络; MATLAB；PID DESIGN ON PLC-BASED NEURAL NETWORK PID CONTROLLER Abstract: PID control is one of the most important control in the control area,. With the artificial neural network research in-depth study, the researchers have also begun to integrate neural networks and PID control, in order to improve and overcome the performance of traditional PID control providing a new way of thinking. This paper introduces a neural network and the PID neural network related concepts, the PID neurons Network model and algorithms of the system introduction.For conventional PID control parameters of greenhouse environment control system exists difficult tuning, large overshoot, adjust a long time, anti-interference ability and poor, the use of neural network system performance advantages of self-learning capability enables the system to find the optimal PID control parameters, and sent directly to the conventional PID controller, has been based on BP neural network tuning PID controllers, which effectively solve the problems of the conventional PID. This paper BP neural network PID control simulation for Using MATLAB software , a simulation program written for a given mutation step response and two cases were simulated. Finally a brief Siemens S7-300 PLC, through its development environment for the preparation of the PID control ladder. Keywords: PLC, BP neural network, MATLAB, PID 目录  HYPERLINK \l _Toc11017_#_Toc11017_ 1 绪论