Complex system anomaly detection via learnable temporal-spatial graph with degradation tendency segmentation
ISA Transactions · 2024-09 · 152 · 156-166
DOI: 10.1016/j.isatra.2024.06.025 ↗Kurz gefasst
Fachartikel in der Zeitschrift ISA Transactions, 2024-09. Verfasst von Qinfeng Han und 3 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory for Manufacturing and Systems Engineering, Xi'an Jiaotong University, Xi'an 710049, PR China. Electronic address: qinfenghan373@…. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 53 Literaturangaben.
Abstract (Originalsprache)
To guarantee the safety and reliability of equipment operation, such as liquid rocket engine (LRE), carrying out system-level anomaly detection (AD) is crucial. However, current methods ignore the prior knowledge of mechanical system itself, and seldom unite the observations with the inherent relation in data tightly. Meanwhile, they neglect the weakness and nonindependence of system-level anomaly which is different from component fault. To overcome above limitations, we propose a separate reconstruction framework using worsened tendency for system-level AD. To prevent anomalous feature being attenuated, we first propose to divide single sample into two equal-length parts along the temporal dimension. And we maximize the mean maximum discrepancy (MMD) between feature segments to force encoders to learn normal features with different distributions. Then, to fully explore the multivariate time series, we model temporal-spatial dependence by temporal convolution and graph attention. Besides, a joint graph learning strategy is proposed to handle prior knowledge and data characteristics simultaneously. Finally, the proposed method is evaluated on two real multi-sensor datasets from LRE and the results demonstrate the effectiveness and potential of the proposed method on system-level AD.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- State Key Laboratory for Manufacturing and Systems Engineering, Xi'an Jiaotong University, Xi'an 710049, PR China. Electronic address: qinfenghan373@163.com.
- State Key Laboratory for Manufacturing and Systems Engineering, Xi'an Jiaotong University, Xi'an 710049, PR China. Electronic address: jlstrive2008@mail.xjtu.edu.cn.
- Science and Technology on Liquid Rocket Engine Laboratory, Xi'an Aerospace Propulsion Institute, Xi'an 710100, China. Electronic address: junw83@163.com.
- State Key Laboratory for Manufacturing and Systems Engineering, Xi'an Jiaotong University, Xi'an 710049, PR China. Electronic address: yancyf2019@163.com.
Zitieren
APA
Han, Q., Chen, J., Wang, J., & Feng, Y. (2024). Complex system anomaly detection via learnable temporal-spatial graph with degradation tendency segmentation. ISA Transactions, 152, 156-166. https://doi.org/10.1016/j.isatra.2024.06.025
Vancouver
Han Q, Chen J, Wang J, Feng Y. Complex system anomaly detection via learnable temporal-spatial graph with degradation tendency segmentation. ISA Transactions. 2024;152:156-166. doi:10.1016/j.isatra.2024.06.025
BibTeX
@article{Han2024,
author = {Han, Qinfeng and Chen, Jinglong and Wang, Jun and Feng, Yong},
title = {Complex system anomaly detection via learnable temporal-spatial graph with degradation tendency segmentation},
journal = {ISA Transactions},
year = {2024},
volume = {152},
pages = {156--166},
doi = {10.1016/j.isatra.2024.06.025}
}
EUVIMED