摘要如今，我国的科技高速地发展，工业愈发的进步。但是在工业生产中，各种问题会出现。主要是它有很强的非线性和滞后性。也经常会发生故障。我们经常可以听到某某化工厂发生爆炸或是其他什么事故。很显然，目前的工业安全技术还不完善。现有的故障检测方法虽然有很多，但是它们很多都不能满足生产中过程监控的要求。无论是什么行业，安全生产永远是第一要务。只有实现了安全生产，才能实现该产业的繁荣昌盛。因此，我们必须完善现有的故障检测方法以及研究出新的故障检测方法，以此来让化工及其他工业过程更加安全。生产过程安全稳定地进行。 70315
前人已经研究出了很多故障检测的方法来减小生产过程中事故的发生，这里我就不再赘述。本文主要研究了神经网络。同时研究了TE过程故障检测方法。我将进行以下的工作:
1.总结现有的故障检测方法，并对各种方法进行了比较。
2.研究RBF神经网络。同时研究使用它做故障检测方法。
3.对基于神经网络的TE过程故障检测系统进行仿真。
4.设计通用的基于神经网络的故障检测系统。
该论文有图8幅，表3个，参考文献20篇。
毕业论文关键词 化工过程 TE 过程 故障检测 RBF神经网络
The Design of Fault Detection of Tennessee Eastman Process Based on Neural Network
Abstract Nowadays, our country's science and technology is developing at a high speed.Industry is becoming more and more progressive.But in industrial production,sorts of problems may arise.This is because it has large nonlinear and hysteresis.Also often occurs fault.We can often hear about an explosion or other accident in a chemical plant.It is clear that the current industrial safety technology is still not perfect.There are many fault detection methods now but these can not meet the requirements of process control in production.No matter in what industry, safety is always the top priority.Only realize the safe production, we can achieve the prosperity of the industry.Therefore, we must improve the existing fault detection methods as well as the study of new fault detection methods, in order to let chemical process more safe and ensure the safety and stability of the production process.
The predecessors have developed many methods of fault detection to reduce the accident, I will not repeat them here.This paper mainly studies neural network as well as studies fault detection methods of it.I will complete the following work:
1. Summarize the existing fault detection methods, and compare the various methods.
2.Study RBF neural network as well as study a fault detection method based on it.
3。complete the simulation of the fault detetion of the TE process which based on neural network.
4.Design general fault detection system based on neural network.
Key words: chemical process Tenessee Eastman process Fault detection
RBF neural network
图清单
图序号 图名称 页码
图2-1 前馈型神经网络模型
图2-2 反馈型神经网络模型
图2-3 RBF神经网络拓扑结构
图3-1 TE过程工艺流程图
图3-2 TE过程 神经网络的TE过程故障检测系统设计:http://www.751com.cn/zidonghua/lunwen_79567.html