Reliability Engineering & System Safety, Год журнала: 2024, Номер 253, С. 110563 - 110563
Опубликована: Окт. 6, 2024
Язык: Английский
Reliability Engineering & System Safety, Год журнала: 2024, Номер 253, С. 110563 - 110563
Опубликована: Окт. 6, 2024
Язык: Английский
Expert Systems with Applications, Год журнала: 2023, Номер 237, С. 121338 - 121338
Опубликована: Авг. 26, 2023
Язык: Английский
Процитировано
113Information Fusion, Год журнала: 2023, Номер 95, С. 1 - 16
Опубликована: Фев. 11, 2023
Язык: Английский
Процитировано
107Journal of Industrial Information Integration, Год журнала: 2023, Номер 33, С. 100469 - 100469
Опубликована: Апрель 27, 2023
Язык: Английский
Процитировано
80Reliability Engineering & System Safety, Год журнала: 2023, Номер 235, С. 109253 - 109253
Опубликована: Март 20, 2023
Язык: Английский
Процитировано
49Engineering Applications of Artificial Intelligence, Год журнала: 2023, Номер 124, С. 106548 - 106548
Опубликована: Июнь 15, 2023
Язык: Английский
Процитировано
44Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 134, С. 108678 - 108678
Опубликована: Июнь 3, 2024
Язык: Английский
Процитировано
36Applied Acoustics, Год журнала: 2024, Номер 217, С. 109807 - 109807
Опубликована: Янв. 4, 2024
Язык: Английский
Процитировано
29Reliability Engineering & System Safety, Год журнала: 2024, Номер 249, С. 110208 - 110208
Опубликована: Май 29, 2024
Язык: Английский
Процитировано
28IEEE Transactions on Instrumentation and Measurement, Год журнала: 2024, Номер 73, С. 1 - 15
Опубликована: Янв. 1, 2024
Bearing fault diagnosis is essential for ensuring the safety and reliability of industrial systems. Recently, deep learning approaches, especially convolutional neural network, have demonstrated exceptional performance in bearing diagnosis. However, limited availability training samples has been a persistent issue, leading to significant reduction diagnostic accuracy. Additionally, noise interference or load variation during operation pose challenges To tackle above issues, this paper explores application quadratic neuron with attention-embedded networks introduces trusted multi-scale strategy that fully considers characteristics vibration signals. Building upon these concepts, network proposed faults Experimental results indicate outperforms six stateof-the-art under superimposed on small samples.
Язык: Английский
Процитировано
21Electronics, Год журнала: 2025, Номер 14(2), С. 341 - 341
Опубликована: Янв. 17, 2025
Rotating machines are vital for ensuring reliability, safety, and operational availability across various industrial sectors. Among the faults that can affect these machines, shaft misalignment is particularly critical due to its impact on other components connected shaft, making it a key focus diagnostic systems. Misalignment lead significant energy losses, therefore, early detection crucial. Vibration analysis an effective method identifying at stage, enabling corrective actions before negatively impacts equipment efficiency consumption. To improve monitoring efficiency, essential system not only intelligent but also capable of operating in real-time. This study proposes methodology diagnosing by combining wavelet transform feature extraction transfer learning fault classification. The accuracy proposed soft real-time solution validated through comparison with time-frequency transformation techniques networks. includes experimental procedure simulating using laser measurement tool. Additionally, evaluates thermal vibration signature each type multi-sensor monitoring, highlighting effectiveness robustness approach. First, used obtain good representation signal domain. step allows features from signals. Then, network processes different layers identify their severity. combination provides decision-support tool faults, monitoring. tested two datasets: first public dataset, while second was created laboratory simulate alignment demonstrate effect this defect imaging. evaluation carried out criteria methodology. results highlight potential implementing faults.
Язык: Английский
Процитировано
4