Tavasolipour, Elham, Hamidi Beheshti, Mohammad Taghi, Ramezani, Amin, Ramezani, Amin (2012). An integrated process monitoring approach combining dynamic independent component analysis and local outlier factor. , 11(4), 1-10.
Elham Tavasolipour; Mohammad Taghi Hamidi Beheshti; Amin Ramezani; Amin Ramezani. "An integrated process monitoring approach combining dynamic independent component analysis and local outlier factor". , 11, 4, 2012, 1-10.
Tavasolipour, Elham, Hamidi Beheshti, Mohammad Taghi, Ramezani, Amin, Ramezani, Amin (2012). 'An integrated process monitoring approach combining dynamic independent component analysis and local outlier factor', , 11(4), pp. 1-10.
Tavasolipour, Elham, Hamidi Beheshti, Mohammad Taghi, Ramezani, Amin, Ramezani, Amin An integrated process monitoring approach combining dynamic independent component analysis and local outlier factor. , 2012; 11(4): 1-10.

An integrated process monitoring approach combining dynamic independent component analysis and local outlier factor

The Modares Journal of Electrical Engineering
Article 1, Volume 11, Issue 4, 2012, Pages 1-10    PDF (831.6 K)
Authors
Elham Tavasolipour1; Mohammad Taghi Hamidi Beheshti* 2; Amin Ramezani3; Amin Ramezani3
1M.s
2Associate Professor Tarbiat Modares University
3Assistant Professor Tarbiat Modares University
Abstract
In this paper a novel process monitoring scheme for reducing the type І and type ІІ error rates in the monitoring phase is proposed. First, the proposed approach uses an augmented data matrix to implement the process dynamic. Then, we apply independent component analysis (ICA) transformation to the augmented data matrix, and eliminate the outliers using the local outlier factor (LOF) algorithm. Finally, the control limit based on the LOF value of the cleaned data are obtained. In the monitoring phase, if the LOF value of each sample exceeds the control limit, fault has occurred; otherwise, data is normal. The proposed method is applied to fault detection in both a simple multivariate dynamic process and the Tennessee Eastman process. In both processes, type І and type ІІ error rates are witnessed to reduce by considering the process dynamic and performing the LOF algorithm. Results clearly indicate better performance of the proposed scheme compared to the alternative methods.
Keywords
Local Outlier Factor; Independent Component Analysis; Tennessee Eastman process; Fault detection
Statistics
Article View: 80
PDF Download: 51
Statistics
Number of Journals45
Number of Issues2,171
Number of Articles24,672
Article View24,384,120
PDF Download17,531,348