Toggle Main Menu Toggle Search

Open Access padlockePrints

Online Process Monitoring through Integration of Joint Recurrence Plot and Convolutional Neural Networks

Lookup NU author(s): Yiran Dong, Dr Jie ZhangORCiD, Dr Chris O'MalleyORCiD

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.


Abstract

© 2024 Elsevier B.V.This paper proposes a new method for online process fault diagnosis through integration of joint recurrence plot (JRP) and convolutional neural networks (CNN). JRP is used to extract features from the major principal component of the process operational data. The extracted features are used as the inputs to a CNN for fault diagnosis. To facility online fault diagnosis, a sliding window of the major principal components is used in generating JRP. The proposed method is demonstrated on a simulated continuous stirred tank reactor (CSTR). The results show that the proposed method can achieve diagnosis accuracy of 99.49% and 97.12% on the training and testing data sets respectively, higher than those of integrating recurrence plot and CNN.


Publication metadata

Author(s): Dong Y, Zhang J, O'Malley C

Publication type: Conference Proceedings (inc. Abstract)

Publication status: Published

Conference Name: 34th European Symposium on Computer Aided Process Engineering /15th International Symposium on Process Systems Engineering (ESCAPE-34/PSE2024)

Year of Conference: 2024

Pages: 1627-1632

Print publication date: 28/06/2024

Online publication date: 26/06/2024

Acceptance date: 02/04/2024

ISSN: 1570-7946

Publisher: Elsevier B.V.

URL: https://doi.org/10.1016/B978-0-443-28824-1.50272-6

DOI: 10.1016/B978-0-443-28824-1.50272-6

Library holdings: Search Newcastle University Library for this item

Series Title: Computer Aided Chemical Engineering

ISBN: 9780443288241


Share