A Novel Integrated Fault Diagnosis Method Based on Digital Twins
Signals · 03.04.2025 · 6 (2) · 18
DOI: 10.3390/signals6020018 ↗Kurz gefasst
Fachartikel in der Zeitschrift Signals, online erschienen am 03.04.2025. Verfasst von Xiangrui Hu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Project of Department of Science and Technology of Hubei Province.
Abstract (Originalsprache)
Fault diagnosis is essential in industrial production. With the advancement of IoT technology, real-time data acquisition and storage have become feasible, enabling deep learning-based fault diagnosis methods to achieve remarkable results. However, existing approaches often overlook the temporal characteristics of fault occurrences and struggle with data imbalance between normal and faulty conditions, impacting diagnostic performance. To address these challenges, this paper proposes an integrated fault diagnosis method that incorporates data balancing, feature extraction, and temporal information analysis. The approach consists of two key components: (1) dataset construction using digital twin technology and (2) an integrated fault diagnosis model (CNN-BLSTM-attention). Digital twin technology generates virtual data under various operating conditions, mitigating the small-sample issue. The proposed model leverages a sliding window mechanism to capture both feature and temporal information, enhancing fault pattern recognition. Experimental results demonstrate that, compared to traditional methods, this approach effectively reduces noise interference and achieves a high diagnostic accuracy of 96.46%, validating its robustness in complex industrial settings. This research provides valuable theoretical and practical insights for improving fault diagnosis in industrial equipment such as screw presses.
Förderung
- Project of Department of Science and Technology of Hubei Province
Zitieren
APA
Hu, X., Liu, L., Quan, Z., Huang, J., & Liu, J. (2025). A Novel Integrated Fault Diagnosis Method Based on Digital Twins. Signals, 6(2), 18. https://doi.org/10.3390/signals6020018
Vancouver
Hu X, Liu L, Quan Z, Huang J, Liu J. A Novel Integrated Fault Diagnosis Method Based on Digital Twins. Signals. 2025;6(2):18. doi:10.3390/signals6020018
BibTeX
@article{Hu2025,
author = {Hu, Xiangrui and Liu, Linglin and Quan, Zhengyu and Huang, Jinguo and Liu, Jing},
title = {A Novel Integrated Fault Diagnosis Method Based on Digital Twins},
journal = {Signals},
year = {2025},
volume = {6},
number = {2},
pages = {18},
doi = {10.3390/signals6020018}
}
EUVIMED