Reconstruction based approach to sensor fault diagnosis using auto-associative neural networks

Abstract

Fault diagnostics is an important research area including different techniques. Principal component analysis (PCA) is a linear technique which has been widely used. For nonlinear processes, however, the nonlinear principal component analysis (NLPCA) should be applied. In this work, NLPCA based on auto-associative neural network (AANN) was applied to model a chemical process using historical data. First, the residuals generated by the AANN were used for fault detection and then a reconstruction based approach called enhanced AANN (E-AANN) was presented to isolate and reconstruct the faulty sensor simultaneously. The proposed method was implemented on a continuous stirred tank heater (CSTH) and used to detect and isolate two types of faults (drift and offset) for a sensor. The results show that the proposed method can detect, isolate and reconstruct the occurred fault properly.

Keywords

fault diagnosis / nonlinear principal component analysis / auto-associative neural networks

References

[1] SubramanianV, RengaswamyR, KavuriN, YinK. A review of process fault detection and diagnosis. Part III: Process history based methods [J]. Computers and Chemical Engineering, 2003, 27: 327-346

[2] ThissenU, MelssenW J, BuydensL M C. Nonlinear process monitoring using bottle-neck neural networks [J]. Analytica Chimica Acta, 2001, 446: 371-383

[3] KourtiT. Process analysis and abnormal situation detection from theory to practice [J]. IEEE Control Systems Magazine, 2002400-405

[4] ZhuQ-x, LiC-fe. Dimensionality reduction with input training neural network and its application in chemical process modeling [J]. Chinese J Chem Eng, 2006, 145: 597-603

[5] Diego Garcia-AlvarezFault detection using principal component analysis (PCA) in a waste water treatment (WWTP) [M], 2010, Valladolid, University of Valladolid Press: 55-59

[6] Jonathon ShlensA tutorial on principal component analysis [D], 2005, San Diego La Jolla, Institute for Nonlinear Science, University of California

[7] HeX-y, LuoH-m, ShaoY-ming. Online fault detection of excavator’s hydraulic system [J]. International Journal of Digital Content Technology and its Applications, 2011, 5(7): 258-260

[8] YangQ-song. Model based and data driven fault diagnosis methods with applications to process monitoring [D]. Case Western Reserve University, 200825-36

[9] MarkA K. Nonlinear principal component analysis using auto-associative neural networks [J]. Alche Journal, 1991, 37(2): 233-243

[10] Mahmoodreza SharifiSensor fault diagnosis using principal component analysis [D], 2009, America, University of Texas A & M

[11] HeQ-h, HeX-y, ZhuJ-xin. Fault detection of excavator’s hydraulic system based on dynamic principal component analysis [J]. Journal of Central South University: Science and Technology, 2008, 15: 700-705

[12] NajafiM, CulpC, LangariR. Performance study of enhanced auto-associative neural networks for sensor fault detection [C]. IEEE Int Conference on Fuzzy Systems. Budapest, 2004453-456

[13] ThornhillN F, SachinC, SirishL S. A continuous stirred tank heater simulation model with applications [J]. Journal of Process Control, 2008, 18: 347-360