Adaptive Detection of Blower Blade Damage Facing Uncertainties in Multisensor Information Fusion
Structural Control and Health Monitoring · 2026-01 · 2026 (1)
DOI: 10.1155/stc/5731551 ↗Kurz gefasst
Fachartikel in der Zeitschrift Structural Control and Health Monitoring, online erschienen am 13.11.2026. Verfasst von Shuyi Chen und 6 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China, Natural Science Foundation of Guangdong Province.
Abstract (Originalsprache)
Intelligent diagnosis for industrial equipment is challenged by multi‐s006Furce data uncertainty and complex topological dependencies between components. Traditional methods struggle to balance dynamic environmental adaptability with modeling intricate relationships. To address these issues, a novel intelligent graph neural network–based damage diagnosis framework integrating acoustic and vibration data is proposed. First, a conflict‐aware D‐S evidence inference module is designed to adaptively adjust acoustic‐vibration fusion weights based on sliding‐window confidence scores and optimize conflicting evidence combination rules. Second, a graph neural network model is constructed, utilizing D‐S‐fused acoustic‐vibration window evidence as nodes and feature dependencies as edges. The spatiotemporal features are aggregated through a graph‐based fine‐grained extraction mechanism to enhance damage‐sensitive region localization capabilities. Finally, a node‐edge level progressive redundancy removal strategy is developed. By jointly evaluating redundant connections and nodes in the graph structure using attention weights and node information gain, redundancy in the graph structure is dynamically thinned to suppress redundant propagation and reduce model complexity. Experiments validated on a self‐built multisensor platform for centrifugal blowers demonstrate that the proposed method significantly outperforms existing models in accuracy. Ablation studies and visualization analyses confirm the effective collaboration among modules and showcase its potential for dynamic diagnostics in industrial settings.
Förderung
- National Natural Science Foundation of China
- National Key Research and Development Program of China
- Natural Science Foundation of Guangdong Province
Zitieren
APA
Chen, S., Wu, W., Zhao, X., Huang, J., Feng, K., Zhuang, J., & Liang, Z. (2026). Adaptive Detection of Blower Blade Damage Facing Uncertainties in Multisensor Information Fusion. Structural Control and Health Monitoring, 2026(1). https://doi.org/10.1155/stc/5731551
Vancouver
Chen S, Wu W, Zhao X, Huang J, Feng K, Zhuang J, et al. Adaptive Detection of Blower Blade Damage Facing Uncertainties in Multisensor Information Fusion. Structural Control and Health Monitoring. 2026;2026(1). doi:10.1155/stc/5731551
BibTeX
@article{Chen2026,
author = {Chen, Shuyi and Wu, Weihai and Zhao, Xiaoli and Huang, Jiale and Feng, Ke and Zhuang, Jichao and Liang, Zhongwei},
title = {Adaptive Detection of Blower Blade Damage Facing Uncertainties in Multisensor Information Fusion},
journal = {Structural Control and Health Monitoring},
year = {2026},
volume = {2026},
number = {1},
doi = {10.1155/stc/5731551}
}
EUVIMED