摘要在工业控制过程中，大多采用传统的 PID 控制方式，其鲁棒性能好、结构简 单、便于实现，但是随着生产工艺的日益复杂和人们对工业总体性能要求的不断 提高，传统的 PID 控制方法已经难以满足工业控制要求。因为传统 PID 控制器 的参数是根据被控对象的数学模型确定的。当被控对象的数学模型是变化的、非 线性的时候，PID 参数不易根据其实际的情况作出调整，达不到控制要求，使控 制系统的控制品质下降。特别是在具有纯滞后特性的工业过程中，传统的 PID 控制方式更难满足控制要求。然而，神经网络作为现代信息处理技术的一种，正 在很多应用中显示了它的优越性，同传统的 PID 控制器相比较，神经网络 PID 控制有许多优点，并且仍将成为未来研究与应用的重点技术之一。72710

本文主要内容是将温室环境中的温度控制问题与 BP 神经网络相结合，我们 已知，被控对象温度存在着参数的不准确性和纯滞后等特性，本文先是分析理论 知识，其次再利用 MATLAB 仿真对 BP 神经网络 PID 控制器进行仿真实验，结 果设计出了满足要求的基于 BP 神经网络控制器。

该设计有图 9 幅，参考文献 22 篇。

毕业论文关键词:神经网络 PID 控制器 温度控制系统

Design method of BP neural network controller for temperature control system

Abstract In the process of industrial control, traditional PID control is the most way of control , its robust performance is good, simple structure, easy to implement, but with the increasing complexity of the production process and the people on the overall industrial performance requirements continue to increase, the traditional PID control method has been difficult to meet the requirements of industrial control。 Because the parameters of traditional PID controller is based on the mathematical model of the controlled object。When the mathematical model  is  variable, non-linear,PID parameters is not easy to adjust according to their actual situation, traditional PID control can not reach the requirements of industrial control, leading to the decrease of quality of control system。Particularly in the industrial process with a pure hysteresis characteristics, the traditional PID control is more difficult to meet the control requirements。 However, as a modern information processing technology,neural network are being demonstrated its superiority in many applications。comparing with the traditional PID controller , neural network PID control has many advantages。 And neural network PID control technology will continue to be one of the research and application of key technologies in the future。

The main contents Greenhouse Environment temperature control problem, there is not a temperature controlled object parameters of accuracy and time characteristics, we use MATLAB simulation of the BP neural network PID controller,indicating good control effect based on BP neural network PID controller。

The paper has 9 pictures,22 references

Key words: neural network PID controller temperature control system

目录

摘要 I

Abstract II

目录 III

图清单 IV

1 概述 1

1。1 选题的背景及意义 1

1。2 温室环境控制技术的国内外发展及研究现状 1

1。3 本文的研究内容及其方法