EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DART-GNN: A Dynamic Recurrent Graph Neural Network for Multivariate Time Series Anomaly Detection

Yuan li, Yuhang Zhou, Qingzhong Yan, Xiang Wu, Yuming Bo

International Journal of Artificial Intelligence and Green Manufacturing · 30.09.2025 · 1 (3)

DOI: 10.67119/212daw81 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Artificial Intelligence and Green Manufacturing, online erschienen am 30.09.2025. Verfasst von Yuan li und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Detecting anomalies in multivariate time series is critical for the safety and reliability of complex cyber-physical systems. Graph Neural Networks (GNNs) have shown great promise in this area by explicitly modeling the relational structure between sensors to improve detection. However, the performance of most GNNs is constrained by their reliance on static graphs, which are unable to capture the evolving nature of relationships between sensors in dynamic environments. To address this limitation, we propose the Dynamic Attention Recurrent Two-step Graph Neural Network (DART-GNN). Our framework constructs a time-specific dependency graph by first using a Gated Recurrent Unit (GRU) to encode temporal context and then applying a self-attention mechanism to infer relationships. A two-step Graph Attention Network (GAT) then performs deep aggregation on this dynamic graph, enabling the model to capture complex, higher-order interactions by propagating information from second-order neighbors. We validated our model's performance through rigorous experiments on the widely-used SWaT and WADI public benchmarks. The results confirm that DART-GNN achieves a new state-of-the-art, demonstrating superior performance compared to a broad spectrum of baseline methods.

Zitieren

APA

li, Y., Zhou, Y., Yan, Q., Wu, X., & Bo, Y. (2025). DART-GNN: A Dynamic Recurrent Graph Neural Network for Multivariate Time Series Anomaly Detection. International Journal of Artificial Intelligence and Green Manufacturing, 1(3). https://doi.org/10.67119/212daw81

Vancouver

li Y, Zhou Y, Yan Q, Wu X, Bo Y. DART-GNN: A Dynamic Recurrent Graph Neural Network for Multivariate Time Series Anomaly Detection. International Journal of Artificial Intelligence and Green Manufacturing. 2025;1(3). doi:10.67119/212daw81

BibTeX
@article{li2025,
  author = {li, Yuan and Zhou, Yuhang and Yan, Qingzhong and Wu, Xiang and Bo, Yuming},
  title = {DART-GNN: A Dynamic Recurrent Graph Neural Network for Multivariate Time Series Anomaly Detection},
  journal = {International Journal of Artificial Intelligence and Green Manufacturing},
  year = {2025},
  volume = {1},
  number = {3},
  doi = {10.67119/212daw81}
}
Kostenlos starten →