Aluminium process fault detection by Multiway Principal Component Analysis

• Published in: Control engineering practice Vol. 19; no. 4; pp. 367 - 379
• Main Authors: Majid, Nazatul Aini Abd, Taylor, Mark P., Chen, John J.J., Stam, Marco A., Mulder, Albert, Young, Brent R.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.04.2011; Elsevier
• Subjects: Aluminium; Aluminium electrolysis; Aluminum; Anode effects; Anode spikes; Applied sciences; Computer science; control theory; systems; Control theory. Systems; Dynamical systems; Dynamics; Electrolysis; Electrometallurgy; Exact sciences and technology; Fault detection; Industrial metrology. Testing; Mechanical engineering. Machine design; Metals. Metallurgy; Modelling and identification; Multiway PCA; Principal component analysis; Process control. Computer integrated manufacturing; Production of metals; Production of non ferrous metals. Process materials; Real time; Real-time fault detection; Smelters; Anode spikes; Aluminium electrolysis; Anode effects; Multiway PCA; Real-time fault detection; Process dynamics; Multivariate process; Aluminium production; Smelting; Continuous time; Aluminium; Electrometallurgy; Real time; Real time system; Continuous process; Electrolysis; Batch process; Failure detection; Alumina; Fault diagnostic; Principal component analysis; Defect detection
• ISSN: 0967-0661; 1873-6939
• DOI: 10.1016/j.conengprac.2010.12.005

Real-time fault detection is difficult to perform in an aluminium smelter because the continuous aluminium electrolysis is operated batchwise in terms of material additions, meaning the measurements obtained from the process are dynamic, multivariate and limited. This paper presents a new framework based on Multiway Principal Component Analysis (MPCA) to detect faults in real-time in the industrial continuous aluminium electrolysis process. This real-time fault detection system incorporates the dynamic behaviour of two important operations in the continuous aluminium electrolysis process, alumina feeding and anode changing. The methodology is demonstrated using real data from an operating aluminium smelter, and is shown to be effective in the early detection of anode spikes and anode effects. ► Detection of anode spikes and anode effects in a real aluminium smelter. ► An alumina feeding cycle was treated as a batch operation using MPCA. ► Monitoring numerous cells using the same reference model.