Abstract
In order to adaptably monitor product qualities during real industrial process, a new multivariate statistical process monitoring scheme combining projection to latent spaces (PLS) and Support Vector Domain Description (SVDD) is proposed. PLS can establish the monitoring space, which maximizes the correlation between process variables and quality variables and enable product qualities monitoring through process variables. SVDD can define the admissible domain by normal operation data without constraints about data distribution. Moreover, with kernel functions it can even provide a tight admissible domain for the operation data. Such characteristics make it suitable for practical production processes. This scheme is then applied to Tennessee Eastman process, and its efficiency for fault detection is proved by introducing simulated process faults. Analysis about its limits in fault detection is also presented.
Sign-in/Register to access full text options 
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.
