Title

改进PCA在发酵过程监测与故障诊断中的应用

Translated Titles

Application of Improved PCA to Fermentation Process Monitoring and Fault Diagnosis

Authors

肖应旺(Ying-Wang Xiao);徐保国(Bao-Guo Xu)

Key Words

主元分析 ; 统计过程监测 ; 发酵 ; 故障诊断 ; PCA ; statistical process monitoring ; fermentation ; fault diagnosis

PublicationName

控制與決策

Volume or Term/Year and Month of Publication

20卷5期(2005 / 05 / 01)

Page #

571 - 574

Content Language

簡體中文

Chinese Abstract

提出一种改进的主元分析(PCA)方法。利用主元相关变量残差统计量代替平方预测误差Q统计量,并采用累积方差贡献率及复相关系数确定PCA模型的主元数。将改进的主元分析法应用于粘菌素发酵过程监测和故障诊断中,仿真结果表明改进的PCA方法避免了Q统计量的保守性,并保证了主元子空间中的信息存量。与一种基于特征子空间的系统性能监控方法相比较,改进的PCA方法具有更强的有效性。

English Abstract

An improved principal component analysis (PCA) is presented which uses principal-component-related variable residual (PVR) statistic to replace Q-statistic. The principal components in PCA is decided by virtue of cumulative percent variance and multi-correlation coefficients. The improved PCA is applied to mycetozoan fermentation process monitoring and fault diagnosis. The simulation result shows that the improved PCA can avoid the conservation of Q-statistical test and ensure enough information in principal component subspace. Compared with a system performance monitoring based on characteristic subspace, the improved PCA is more effective.

Topic Category 基礎與應用科學 > 資訊科學
工程學 > 機械工程
工程學 > 電機工程