Title: Fault diagnosis method based on robust canonical variate analysis
Abstract: A fault diagnosis method based on robust canonical variate analysis is presented to analyze model data with outliers in industrial processes.This method replaces the traditional correlation coefficient with its robust estimator,and applies projection pursuit technique based on particle swarm optimization algorithm to calculate the canonical variate that maximizes the robust correlation coefficient.Then statistical model is built and monitoring statistics are constructed to detect process faults.The simulation results on a continuous stirred tank reactor(CSTR) system show that robust canonical variate analysis can built accurate statistical model from process data with outliers and monitor process changes more effectively than canonical variate analysis.
Publication Year: 2008
Publication Date: 2008-01-01
Language: en
Type: article
Access and Citation
Cited By Count: 1
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot