基于SVR的非线性系统故障诊断研究  

Fault Diagnosis of Nonlinear Systems Based on Support Vector Regression

作　　者：胡良谋[1] 曹克强[1] 王文栋[2] 徐浩军[1] 董新民[1] 

机构地区：[1]空军工程大学工程学院 [2]94287部队

出　　处：《机械科学与技术》2010年第2期225-228,233,共5页Mechanical Science and Technology for Aerospace Engineering

基　　金：中国博士后科学基金项目资助

摘　　要：针对非线性系统辨识建模和故障诊断难的问题,利用回归型支持向量机(support vector regression,SVR)分别设计了非线性系统的辨识建模系统和故障诊断系统,最后以某一非线性系统为例进行了仿真试验研究,建立了该非线性系统的SVR辨识模型,在此基础上进行了三种典型故障的诊断试验,仿真试验结果验证了该方法的有效性和先进性。Aiming at the difficulty of system identification modeling and fault diagnosis for nonlinear systems,a identification modeling system and a fault diagnosis system are designed for nonlinear systems,respectively,by using support vector regression （SVR） . Simulation of one nonlinear system is carried out as an example. The SVR identification model of the nonlinear system is established. Three typical fault simulation tests are carried out based on the SVR identification model. The simulation results show that the method mentioned above is valid and advanced.

关 键 词：回归型支持向量机(SVR) 非线性系统 系统辨识 故障诊断 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]