Transfer learning with 2D vibration images for fault diagnosis of bearings under variable speed.

Ahmad, Zahoor; Hasan, Md Junayed; Kim, Jong-Myon

Abstract

One of the most critical assignments in fault diagnosis is to decide the finest set of features by evaluating the statistical parameters of the time-domain signals. However, these parameters are vulnerable under variable speed conditions, i.e., different loads, and speeds to capture the dynamic attributes of various health types. Therefore, this paper proposes a vibration imagining-based diagnosis approach for bearing under variable speed conditions. First, a Discrete Cosine Stockwell Transformation (DCST) coefficient-based preprocessing step is proposed to create an identical health pattern for variable speed conditions. Then, from that 2D coefficient matrix, a vibration image is created to capture those health patterns into grayscale. Finally, a Transfer Learning embedded Convolutional Neural Network (TL-CNN) is proposed to inspect the comprehensive structure of the 2D vibration images for final classification. The experimental results show that the proposed method achieved 100% classification accuracy on a publicly available dataset.

Citation

AHMAD, Z., HASAN, M.J. and KIM, J.-M. 2022. Transfer learning with 2D vibration images for fault diagnosis of bearings under variable speed. In Abraham, A., Gandhi, N., Hanne, T., Hong, T.-P., Rios, T.N. and Ding, W. (eds.) Intelligent systems design and applications: proceedings of 21st International conference on intelligent systems design and applications (ISDA 2021), 13-15 December 2021, [virtual event]. Lecture notes in networks and systems, 418. Cham: Springer [online], pages 154-164. Available from: https://doi.org/10.1007/978-3-030-96308-8_14

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 21st International conference on intelligent systems design and applications (ISDA 2021)
• Start Date: Dec 13, 2021
• End Date: Dec 15, 2021
• Acceptance Date: Nov 5, 2021
• Online Publication Date: Mar 27, 2022
• Publication Date: Mar 27, 2022
• Deposit Date: Nov 3, 2022
• Publicly Available Date: Mar 28, 2024
• Publisher: Springer
• Peer Reviewed: Peer Reviewed
• Pages: 154-164
• Series Title: Lecture notes in networks and systems (LNNS)
• Series Number: 418
• Series ISSN: 2367-3389
• Book Title: Intelligent systems design and applications: proceedings of 21st International conference on intelligent systems design and applications (ISDA 2021), 13-15 December 2021
• ISBN: 9783030963071
• DOI: https://doi.org/10.1007/978-3-030-96308-8_14
• Keywords: Bearing; Condition monitoring; Convolutional neural network; Stockwell transformation; Transfer learning
• Public URL: https://rgu-repository.worktribe.com/output/1799360
• Additional Information: In addition to this record, Md Junayed Hasan has also authored the following chapter from the same book title:; AHMAD, Z., HASAN, M.J. and KIM, J.-M. 2022. Centrifugal pump fault diagnosis using discriminative factor-based features selection and K-nearest neighbors. In Abraham, A., Gandhi, N., Hanne, T., Hong, T.-P., Rios, T.N. and Ding, W. (eds.) Intelligent systems design and applications: proceedings of 21st International conference on intelligent systems design and applications (ISDA 2021), 13-15 December 2021, [virtual event). Lecture notes in networks and systems, 418. Cham: Springer [online], pages 145-153. Available from: https://doi.org/10.1007/978-3-030-96308-8_13