Plant‐wide processes monitoring and fault tracing based on causal graphical model
IET Control Theory & Applications · 2024-11 · 18 (17) · 2322-2334
DOI: 10.1049/cth2.12499 ↗Kurz gefasst
Fachartikel in der Zeitschrift IET Control Theory & Applications, online erschienen am 06.10.2023. Verfasst von Xiaolu Chen und 2 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Plant‐wide processes are usually characterized by large scale, multiple operating units and complicated interactions. Effective monitoring for such processes is imperative and challenging. Traditional data‐driven methods have some limitations due to the neglect of internal relationships between operating units. This paper proposes a plant‐wide monitoring and diagnostic framework based on the multi‐variate statistical analysis and causal graphical inference. Initially, the optimized process decomposition is performed by combining the mechanistic knowledge and historical data from the perspective of improving the monitoring performance. Taking into account the fact that shared variables among the different subsystems lead to the information interaction rather than being independent as in the existing methods, the multi‐variate causal model based on probability density estimation is established to identify the quantitative association of the process variables in a single subsystem. The complete model is structured by the link of shared variables. Finally, system anomalies are detected by changes in the probability density of the observed variables; the root cause is pinpointed by the causal inference. Experiments with the Tenessee Eastman (TE) process and Panamax bulk carriers demonstrate the applicability of the proposed methodology.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Chen, X., Yang, Y., & Wang, J. (2024). Plant‐wide processes monitoring and fault tracing based on causal graphical model. IET Control Theory & Applications, 18(17), 2322-2334. https://doi.org/10.1049/cth2.12499
Vancouver
Chen X, Yang Y, Wang J. Plant‐wide processes monitoring and fault tracing based on causal graphical model. IET Control Theory & Applications. 2024;18(17):2322-2334. doi:10.1049/cth2.12499
BibTeX
@article{Chen2024,
author = {Chen, Xiaolu and Yang, Ying and Wang, Jing},
title = {Plant‐wide processes monitoring and fault tracing based on causal graphical model},
journal = {IET Control Theory & Applications},
year = {2024},
volume = {18},
number = {17},
pages = {2322--2334},
doi = {10.1049/cth2.12499}
}
EUVIMED