Title: Nonlinear system monitoring using multiscaled principal components analysis based on neural network
Authors: Hanen Chaouch Jebril; Khaled Ouni; Lotfi Nabli
Addresses: National Engineers School of Monastir, Ibn Al Jazzar City, 5019 Monastir, Tunisia ' National Engineers School of Monastir, Ibn Al Jazzar City, 5019 Monastir, Tunisia ' National Engineers School of Monastir, Ibn Al Jazzar City, 5019 Monastir, Tunisia
Abstract: In this paper, we propose a new method based on multiscaled principal component analysis for nonlinear systems analysis. We introduce nonlinear PCA based on neural networks and discrete wavelet transform. The data matrix describing a nonlinear process is decomposed into five wavelet resolution levels. The neural PCA is applied to each coefficient of details and approximations; we select only the scales having a defect to reconstruct the data matrix. Neural PCA is again applied to the new matrix to determine the defective variables, which are detected using the square predictive error (SPE) statistic and identified using the contributions calculation method. This method is applied to a biological process and shows efficient results.
Keywords: multiscaled analysis; principal component analysis; PCA; neural networks; nonlinear systems; defected variables; system monitoring; discrete wavelet transform; DWT; square predictive error; SPE; biological processes.
International Journal of Modelling, Identification and Control, 2017 Vol.27 No.1, pp.68 - 73
Available online: 20 Feb 2017 *Full-text access for editors Access for subscribers Purchase this article Comment on this article