Autoencoder Based Residual Generation for Fault Detection of Quadruple Tank System

Asgeir Daniel Hallgrimsson, Hans Henrik Niemann, Morten Lind

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

Increasing complexity of industrial processes has made statistical methods for process monitoring and diagnosis a more attractive alternative to model-based methods. A primary reason is that statistical approaches can be formulated to rely less on process knowledge. Since multivariable processes can exhibit complex, nonlinear dynamics, there is a need for methods capable of diagnosing nonlinear process data. A Monte Carlo simulation was conducted on a numerical model of the quadruple tank process (QTP) - a novel multivariate nonlinear process. The simulation was designed so that the QTP exhibited bipartite nonlinear behavior. Reference data obtained from the simulation was used to obtain principal component analysis (PCA) and autoencoder (AE) models. The models generated residuals that were used to monitor the condition of the process. The results showed that AEs, which have nonlinear functionalities, performed better than PCA models at generating residuals.

Original languageEnglish
Title of host publicationProceedings of the 3rd IEEE Conference on Control Technology and Applications
PublisherIEEE
Publication dateAug 2019
Pages994-999
Article number8920588
ISBN (Electronic)9781728127675
DOIs
Publication statusPublished - Aug 2019
Event3rd IEEE Conference on Control Technology and Applications, CCTA 2019 - Hong Kong, China
Duration: 19 Aug 201921 Aug 2019

Conference

Conference3rd IEEE Conference on Control Technology and Applications, CCTA 2019
Country/TerritoryChina
CityHong Kong
Period19/08/201921/08/2019
SponsorCity University of Hong Kong, Hong Kong Automatic Control Association, IEEE

Fingerprint

Dive into the research topics of 'Autoencoder Based Residual Generation for Fault Detection of Quadruple Tank System'. Together they form a unique fingerprint.

Cite this