发布时间 : 星期一 文章基于BP神经网络PID整定原理和算法步骤-精品更新完毕开始阅读
摘 要
神经网络作为一门新兴的信息处理科学，是对人脑若干基本特性的抽象和模拟。它是以人的大脑工作模式为基础，研究自适应及非程序的信息处理方法。这种工作机制的特点表现为通过网络中大量神经元的作用来体现自身的处理功能，从模拟人脑的结构和单个神经元功能出发，达到模拟人脑处理信息的目的。
目前，在国民经济和国防科技现代化建设中神经网络具有广阔的应用领域和发展前景，其应用领域主要表现在信息领域、自动化领域、工程领域和经济领域等。
本文以BP神经网络作为研究对象。研究的内容主要有：首先介绍了神经网络的概念、控制结构，学习方式等。其次，介绍了人工神经元模型，并对BP神经网络的基本原理及推导过程进行详细阐述。再次将BP神经网络的算法应用于PID中，介绍了基于BP神经网络PID整定原理和算法步骤。最后利用 MATLAB/Simulink对BP神经网络PID控制系统进行仿真，得出BP神经网络的控制效果明显好，它具有很强的自整定，自适应功能。
关键词：BP算法，PID控制，自整定
I
ABSTRACT
As a kind of emerging information processing science，the neural network can simulate some basic characteristic of human brain. It is an information-processed method which takes person's cerebrum working pattern as a foundation and studies the model of adaptive and non- program. The characteristics of this kind of work mechanism are that it can show its processing function through the massive neurons function in the network. Then, it starts with simulating the human brain structure and the single neuron function to achieve the goal that simulates the human brain to process information.
Nowadays, the neural network has wide application fields and prospects in the national economy and modernization of national defense science. It mainly applies in information, automation, economical and so on.
This article takes the BP neural network as the research object. The content of the research mainly contain: firstly, it introduces the concept of neural network, control structure and mode of study and so on. Secondly, it introduces the artificial neuron model, the basic principles of BP neural network and the derivation process in detail. Then, it applies BP neural network in the PID, and introduces the tuning principles of PID based the BP neural network and steps of the algorithm. Finally, Matlab/Simulink is used to simulate the BP neural network PID control system. In the consequence, the performance of BP neutral network control significantly good. BP neural network control system has a strong self-tuning, adaptive function.
KEY WORDS: BP algorithm, PID control, self-tuning
II
目 录
摘 要 .......................................................... I ABSTRACT .......................................................... II 第1章 绪论 ........................................................ 2 1.1选题背景和意义 ............................................... 2 1.2神经网络技术国内外发展现状 ................................... 3 第2章 神经网络的原理和应用 ........................................ 6 2.1神经网络的基本概念 ........................................... 6 2.2神经网络的控制结构 ........................................... 6
2.2.1 前馈网络 .............................................. 6 2.2.2 反馈网络 .............................................. 6 2.3神经网络的功能 ............................................... 7 2.4神经网络的学习 ............................................... 7
2.4.1神经网络的学习方式 .................................... 7 2.4.2神经网络的学习算法 .................................... 8 2.5人工神经元（MP）模型 ......................................... 8 2.6 BP算法原理 .................................................. 10 2.7 BP网络的前馈计算 ............................................. 11 2.8 BP网络权系数的调整规则 ...................................... 12 2.9 BP网络学习算法的计算步骤 .................................... 14 2.10本章小结 ................................................... 14 第3章 BP神经网络PID控制方法研究 ................................ 15 3.1引言 ........................................................ 15 3.2 基于BP神经网络的PID整定原理 ............................... 15 3.3 本章小结 .................................................... 19 第4章 仿真研究 ................................................... 20 4.1 BP神经网络自整定PID控制系统 ................................ 20 4.2 仿真结果分析 ................................................ 27 4.3 本章小结 .................................................... 27 第5章 结论与展望 ................................................. 28 参 考 文 献 ....................................................... 30 附 录 .......................................................... 31 致 谢 .......................................... 错误！未定义书签。
1