Wavelets and non-linear principal components analysis for process monitoring

被引：57

作者：

Shao, R ^{[1
]}

Jia, F ^{[1
]}

Martin, EB ^{[1
]}

Morris, AJ ^{[1
]}

机构：

[1] Newcastle Univ, Ctr Proc Analyt & Control Technol, Newcastle Upon Tyne NE1 7RU, Tyne & Wear, England

来源：

CONTROL ENGINEERING PRACTICE | 1999年 / 7卷 / 07期

关键词：

fault detection; smoothing filters; neural networks; multivariate quality control; chemical industry;

D O I：

10.1016/S0967-0661(99)00039-8

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A non-linear principal component analysis (PCA) algorithm is proposed For process performance monitoring based upon an input-training neural network. Prior to assessing the capabilities of the monitoring scheme on an industrial dryer, the data is first pre-processed to remove noise and spikes through wavelet de-noising. The wavelet coefficients obtained are used as the inputs for the non-linear PCA algorithm. Performance monitoring charts with non-parametric control limits are then applied to identify the occurrence of non-conforming operation prior to interrogating differential contribution plots to help identify the potential source of the fault. Encouraging results were achieved. (C) 1999 Elsevier Science Ltd. All rights reserved.

引用

页码：865 / 879

页数：15

共 29 条

[1]

Bishop C. M., 1995, NEURAL NETWORKS PATT

[2]

BOWMAN AW, 1984, BIOMETRIKA, V71, P353

[3]

Daubechies I., 1992, Ten Lectures on Wavelets, DOI 10.1137/1.9781611970104

[4] Nonlinear principal component analysis - Based on principal curves and neural networks [J].