Abstract
Partial least squares (PLS) and linear regression methods have been widely utilized for quality-related fault detection in industrial processes recently. Since these traditional approaches assume that process variables follow Gaussian distribution approximately, their effectiveness will be challenged when facing non-Gaussian processes. To deal with this difficulty, a new quality relevant process monitoring approach based on improved independent component regression (IICR) is presented in this article. Taking high-order statistical information into account, ICA is performed onto process data to produce independent components (ICs). In order to remove irrelevant variation orthogonal to quality variable and keep as much quality-related fault information as possible, a new quality-related independent components selection method is applied to these ICs. Then the regression relationship between filtered ICs and the product quality is built. QR decomposition for regression coefficient matrix is able to give out quality-related and quality-unrelated projectors. After the measured variable matrix is divided into quality relevant and quality irrelevant parts, novel monitoring indices are designed for fault detection. finally, applications to two simulation cases testify the effectiveness of our proposed quality-related fault detection method for non-Gaussian processes.
Highlights
Quality relevant process monitoring has attracted a lot of attention recently
Thanks to the rapid development of automation technology and computers, a large number of process variables can be measured and the corresponding data can be stored, which makes it proper to use data-driven methods for fault detection [2], [3]. When it comes to quality-related process monitoring, partial least squares (PLS) and some regression approaches are popular for the reason that a multivariate statistical model considering of the relationship between measured variables and product quality needs to be constructed
To demonstrate the benefit of orthogonal signal correction (OSC) based quality related independent components filtering method,and that is what we focus on in this article, another faulty data set is generated
Summary
Quality relevant process monitoring has attracted a lot of attention recently. Instead of focusing on the whole process operation status, including the running state of actuators, controllers and sensors, quality-related process monitoring only considering the quality of products [1]. If process variables obey non-Gaussian data distribution, the monitoring statistics in PLS and linear regression methods will miss the higher-order statistical information. When using orthogonal signal correction and orthogonal projections to latent structures as preprocessing tools, the variation which is orthogonal to output Y is removed from input space X This part of X has no value for quality-related fault detection but it contains useful fault information for quality-unrelated monitoring. Independent component regression (ICR) is utilized in this paper to develop a new fault detection approach for quality-relevant non-Gaussian process monitoring. IMPROVED ICR FOR QUALITY RELEVANT FAULT DETECTION The proposed IICR is described in detail, which includes OSC-based independent components filtering, ICR monitoring scheme, and quality variables prediction and monitoring
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
