| 一种非线性非高斯随机系统的故障诊断方法 |
| |
| 引用本文： | 张柏,陈敏泽,周东华.一种非线性非高斯随机系统的故障诊断方法[J].控制工程,2005(Z2). |
| |
| 作者姓名： | 张柏 陈敏泽 周东华 |
| |
| 作者单位： | 清华大学自动化系 北京100084
(张柏，陈敏泽)，清华大学自动化系 北京100084(周东华) |
| |
| 基金项目： | 国家自然科学基金资助项目(60234010)
国家973计划资助项目(2002CB312200) |
| |
| 摘 要： | 针对非线性非高斯随机系统在线故障诊断的问题,运用粒子滤波器提出了一种基于方差自适应粒子滤波器的非线性非高斯随机系统的故障诊断方法,可以用来解决系统的参数偏差型故障诊断问题。通过对连续搅拌釜式反应器(CSTR)的仿真研究,可以看出,该算法实现简单,易于对系统进行在线估计,对于发生缓变和突变的参数偏差型故障的检测与估计均较为有效。
|
| 关 键 词： | 故障诊断 粒子滤波器 序列蒙特卡罗方法 方差自适应 非线性非高斯系统 |
Fault Diagnosis for Non-linear/Non-Gaussian Stochastic Systems |
| |
| ZHANG Bai,CHEN Min-ze,ZHOU Dong-hua.Fault Diagnosis for Non-linear/Non-Gaussian Stochastic Systems[J].Control Engineering of China,2005(Z2). |
| |
| Authors: | ZHANG Bai CHEN Min-ze ZHOU Dong-hua |
| |
| Affiliation: | ZHANG Bai,CHEN Min-ze,ZHOU Dong-hua Department of Automation,Tsinghua University,Beijing 100084,China |
| |
| Abstract: | Most systems in engineering are non-linear and non-Gaussian. The online fault diagnosis of such non-linear/non-Gaussian systems is a problem of great practical importance. In the past few years, particle filters have been developed as an effective tool to solve filtering problems, of non-linear/non-Gaussian stochastic systems. A new method for fault diagnosis of non-linear and non-Gaussian stochastic systems based on variance-adaptive particle filters is introduced.This method is able to detect and estimate fault of parameter bias. A simulation example of continuous stirred tank reactor (CSTR) is presented, which shows the effectiveness of this method. |
| |
| Keywords: | fault diagnosis particle filter sequential Monte Carlo methods variance adaptive non-linear/non-Gaussian systems |
| 本文献已被 CNKI 等数据库收录！ |
