Nonlinear partial least squares with Hellinger distance for nonlinear process monitoring

Harrou, Fouzi; Madakyaru, Muddu; Sun, Ying

Nonlinear partial least squares with Hellinger distance for nonlinear process monitoring

Files

Nonlinear Partial Least Squares with Hellinger Distance.pdf (510.11 KB)

Type

Conference Paper

Authors

Harrou, Fouzi

Madakyaru, Muddu
Sun, Ying

KAUST Department

Applied Mathematics and Computational Science Program
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Statistics Program

KAUST Grant Number

OSR-2015-CRG4-2582

Online Publication Date

2017-02-16

Print Publication Date

2016-12

Date

2017-02-16

Abstract

This paper proposes an efficient data-based anomaly detection method that can be used for monitoring nonlinear processes. The proposed method merges advantages of nonlinear projection to latent structures (NLPLS) modeling and those of Hellinger distance (HD) metric to identify abnormal changes in highly correlated multivariate data. Specifically, the HD is used to quantify the dissimilarity between current NLPLS-based residual and reference probability distributions. The performances of the developed anomaly detection using NLPLS-based HD technique is illustrated using simulated plug flow reactor data.

Citation

Harrou F, Madakyaru M, Sun Y (2016) Nonlinear partial least squares with Hellinger distance for nonlinear process monitoring. 2016 IEEE Symposium Series on Computational Intelligence (SSCI). Available: http://dx.doi.org/10.1109/SSCI.2016.7849878.

Acknowledgements

This publication is based upon work supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No: OSR-2015-CRG4-2582.