Paper Info
Title
Fault detection of multimode non-Gaussian dynamic process using dynamic Bayesian independent component analysis.
Abstract
Independent component analysis (ICA) has been widely used in non-Gaussian multivariate process monitoring. However, it assumes only one normal operation mode and omits the dynamic characteristic of process data. In order to overcome the shortcomings of traditional ICA based fault detection method, an improved ICA method, referred to as dynamic Bayesian independent component analysis (DBICA), is proposed to monitor the multimode non-Gaussian dynamic process. In this method, matrix dynamic augmentation is applied to extract dynamic information from original data. Then for analyzing multimode non-Gaussian data, Bayesian inference and ICA are combined to establish a probability mixture model. The ICA model parameters are obtained by the iterative optimization algorithm and the mode of each observation is determined by Bayesian inference simultaneously. Lastly case studies on one continuous stirring tank reactor (CSTR) simulation system and the Tennessee Eastman (TE) benchmark process are used to demonstrate the effectiveness of the proposed method.
Year
DOI
Venue
2016
10.1016/j.neucom.2016.03.015
Neurocomputing
Keywords
Field
DocType
Multimode process,Dynamic data,Independent component analysis,Bayesian inference,Probability mixture model
Bayesian inference,Pattern recognition,Multivariate statistics,Fault detection and isolation,Computer science,Gaussian,Dynamic data,Artificial intelligence,Independent component analysis,Machine learning,Mixture model,Bayesian probability
Journal
Volume
Issue
ISSN
200
C
0925-2312
Citations 
PageRank 
References 
5
0.44
16
Authors
2
Name
Order
Citations
PageRank
Ying Xu1629.43
Deng Xiaogang211517.49