Localized Fisher discriminant analysis based complex chemical process monitoring

• Published in: AIChE journal Vol. 57; no. 7; pp. 1817 - 1828
• Main Author: Yu, Jie
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 01.07.2011; Wiley; American Institute of Chemical Engineers
• Subjects: Algorithms; Applied sciences; Chemical engineering; chemical process monitoring; Classification; Clusters; Construction; Discriminant analysis; Exact sciences and technology; fault detection and classification; Faults; Gaussian mixture model; localized Fisher discriminant analysis; Mathematical models; Monitoring; Monitoring methods; Monitoring systems; Normal distribution; Preserves; Principal components analysis; Safety; stationarity testing; Tennessee Eastman process; USA, Tennessee; chemical process monitoring; localized Fisher discriminant analysis; Sensitivity analysis; Independent component analysis; Extract; stationarity testing; Algorithm; Modeling; Steady state; Gaussian mixture model; fault detection and classification; Surveillance; Tennessee Eastman process; Failure detection; Partial least squares; Principal component analysis

Complex chemical process is often corrupted with various types of faults and the fault‐free training data may not be available to build the normal operation model. Therefore, the supervised monitoring methods such as principal component analysis (PCA), partial least squares (PLS), and independent co...