文章基本信息

标题：Quality-related Fault Detection Approaches Based on Data Preprocessing * * This work was supported by National Natural Science Foundations of China (No. 61503039, No. 61503040)
本地全文：下载
作者：Guang Wang ; Jianfang Jiao ; Shen Yin 等
期刊名称：IFAC PapersOnLine
印刷版ISSN：2405-8963
出版年度：2017
卷号：50
期号：1
页码：15740-15747
DOI：10.1016/j.ifacol.2017.08.2305
语种：English
出版社：Elsevier
摘要：AbstractThis work focus on the issue of quality-related fault detection. A new idea of data-preprocessing is proposed with two specific instances are designed based on orthogonal signal correction (OSC) and orthogonal projections to latent structures (OPLS). Different from existing results, the new idea allows directly designing test statistics in the data subspaces obtained from data preprocessing without building any linear regression model like partial least squares (PLS). Benefit from such a direct feature, the designed new methods are more simple in engineering implementation and their performances are also more stable than conventional approaches. Simulation results on a widely used literature example and an industrial example demonstrate the effectiveness of the proposed new methods.
关键词：KeywordsData-drivenfault detectionquality-relateddata preprocessing