Modeling and monitoring for handling nonlinear dynamic processes

Authors:
Yingwei Zhang;Jiayu An;Zhiming Li;Hong Wang
Affiliations:
State Laboratory of Synthesis Automation of Process Industry, Northeastern University, Shenyang, Liaoning 110004, PR China;State Laboratory of Synthesis Automation of Process Industry, Northeastern University, Shenyang, Liaoning 110004, PR China;State Laboratory of Synthesis Automation of Process Industry, Northeastern University, Shenyang, Liaoning 110004, PR China;Univ. Manchester, Control Syst. Ctr., Sch. Elect. & Elect. Engn., Manchester M60 1QD, Lancs, England, UK
Venue:
Information Sciences: an International Journal
Year:
2013

Citing 4
Cited 0

Complex system fault diagnosis based on a fuzzy robust wavelet support vector classifier and an adaptive Gaussian particle swarm optimization

Information Sciences: an International Journal
A novel ant-based clustering algorithm using the kernel method

Information Sciences: an International Journal
Scaling the kernel function based on the separating boundary in input space: A data-dependent way for improving the performance of kernel methods

Information Sciences: an International Journal
An adaptive learning algorithm for principal component analysis

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.07

Visualization

Abstract

In this paper, a new online monitoring approach is proposed for handling the dynamical multimode problem in the industrial processes. The contributions are as follows: (1) extracting method of the common characteristics from different modes is proposed; (2) nonlinear dynamic monitoring method is proposed; and (3) a new model analysis method is proposed. There are both similarity and dissimilarity in the underlying correlations of different modes. After two different subspaces are separated, models of the common and specific subspaces are built respectively. Then the common subspace and specific subspace are analyzed, where the monitoring process is carried out in each subspace. When the mode switches, the specific monitoring model is changed. The corresponding confidence regions are constructed according to their models respectively. The effectiveness of the proposed method has been demonstrated via simulated examples.