摘要
介绍了一种非线性故障检测方法———核主元分析法(KPCA),通过核函数来完成非线性变换,将变量由非线性的输入空间转换到线性的特征空间.在特征空间中使用PCA计算主元,构造T2和SPE统计量检测过程故障的发生.提出了一种KPCA贡献图计算方法,根据测量变量和非线性主元的相关性,计算测量变量的贡献量绘制贡献图,用于故障变量的分离.仿真结果表明,KPCA方法可以比PCA方法更加迅速的检测到故障的发生,利用KPCA贡献图可以较好的辨识出故障变量.
A nonlinear fault detection method based on kernel principal component analysis (KPCA) is introduced. KPCA performs nonlinear transformation by kernel function to map the nonlinear input space into linear feature space.Based on T^2 and SPE charts in feature space, principal component analysis(PCA)can be used to detect faults. KPCA contribution plots are proposed to isolate faulty variables. According to correlation between measured variables and nonlinear principal components, the contribution of each variable is calculated to give contribution plots. The simulation results indicate that KPCA is superior to PCA and that contribution plots can isolate the faulty variables well.
出处
《山东大学学报（工学版）》
CAS
2005年第3期103-106,共4页
Journal of Shandong University（Engineering Science）
基金
国家863计划资助项目(编号:2004AA412050)
关键词
核主元分析法
贡献图
非线性过程
故障检测
故障诊断
kernel principal component analysis
contribution plots
nonlinear process
fault detection
fault diagnosis
