EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Dynamic Temporal Denoise Neural Network with Multi-Head Attention for Fault Diagnosis Under Noise Background

Zhongzhi Li iD, Rong Fan, Jinyi Ma iD, Jianliang Ai, Yiqun Dong

Sensors · 23.10.2024 · 24 (21) · 6813

DOI: 10.3390/s24216813 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 23.10.2024. Verfasst von Zhongzhi Li und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Aeronautics and Astronautics, Fudan University, Shanghai 056004, China. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shanghai Sailing Program, Natural Science Foundation of Shanghai, Shanghai Pujiang Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Fault diagnosis plays a crucial role in maintaining the operational safety of mechanical systems. As intelligent data-driven approaches evolve, deep learning (DL) has emerged as a pivotal technique in fault diagnosis research. However, the collected vibrational signals from mechanical systems are usually corrupted by unrelated noises due to complicated transfer path modulations and component coupling. To solve the above problems, this paper proposed the dynamic temporal denoise neural network with multi-head attention (DTDNet). Firstly, this model transforms one-dimensional signals into two-dimensional tensors based on the periodic self-similarity of signals, employing multi-scale two-dimensional convolution kernels to extract signal features both within and across periods. Secondly, for the problem of lacking denoising structure in traditional convolutional neural networks, a temporal variable denoise (TVD) module with dynamic nonlinear processing is proposed to filter the noises. Lastly, a multi-head attention fusion (MAF) module is used to weight the denoted features of signals with different periods. Evaluation on two datasets, Case Western Reserve University bearing dataset (single sensor) and Real aircraft sensor dataset (multiple sensors), demonstrates that the DTDNet can reduce the useless noises in signals and achieve a remarkable improvement in classification performance compared with the state-of-the-art method. DTDNet provides a high-performance solution for potential noise that may occur in actual fault diagnosis tasks, which has important application value.

Schlagwörter

Stichwörter der Autor:innen

fault diagnosismixed precision calculationmulti-time resolution decompositionnonlinear denoisingshort-time fourier transform

Autor:innen und Institutionen

  • Department of Aeronautics and Astronautics, Fudan University, Shanghai 056004, China.

ORCID-Profile: Zhongzhi Li, Jinyi Ma

Förderung

  • Shanghai Sailing Program
  • Natural Science Foundation of Shanghai
  • Shanghai Pujiang Program
  • Yiqun Dong: Shanghai Sailing Program

Zitieren

APA

Li, Z., Fan, R., Ma, J., Ai, J., & Dong, Y. (2024). Dynamic Temporal Denoise Neural Network with Multi-Head Attention for Fault Diagnosis Under Noise Background. Sensors, 24(21), 6813. https://doi.org/10.3390/s24216813

Vancouver

Li Z, Fan R, Ma J, Ai J, Dong Y. Dynamic Temporal Denoise Neural Network with Multi-Head Attention for Fault Diagnosis Under Noise Background. Sensors. 2024;24(21):6813. doi:10.3390/s24216813

BibTeX
@article{Li2024,
  author = {Li, Zhongzhi and Fan, Rong and Ma, Jinyi and Ai, Jianliang and Dong, Yiqun},
  title = {Dynamic Temporal Denoise Neural Network with Multi-Head Attention for Fault Diagnosis Under Noise Background},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {21},
  pages = {6813},
  doi = {10.3390/s24216813}
}
Kostenlos starten →