Journal of Engineering, Project, and Production Management, 2026, 16(3), 2026-269
Rolling Bearing Fault Detection Based on Deep Residual Capsule Network
Undergraduate Student, College of Mechanical and Electrical Engineering, Northeast Forestry University, Harbin, 150040, China, E-mail: Yuzhouzhangzyz@outlook.com
Production Management
Received February 27, 2026; revised April 16, 2026; accepted April 18, 2026
Available online May 4, 2026
Abstract: Bearing failure can cause equipment to stop running, reduce production efficiency, and in severe cases, even lead to safety accidents. Traditional bearing fault detection methods often lack sufficient feature extraction, limited generalization, and poor anti-noise performance. In this sense, this paper proposes a bearing failure detection method based on a Deep Residual Capsule Network (DRCN). DRCN combines the advantages of the Deep Residual Network (ResNet) and the Capsule Network (CapsNet). ResNet, by leveraging its remaining structure, can automatically learn deep fault features and achieve high robustness, thereby avoiding the limitations of manual feature extraction and alleviating the problem of network degradation. CapsNet captures the spatial hierarchical relationship between features through a dynamic routing mechanism, effectively improving the deficiency of traditional convolutional neural networks in expressing feature direction and position information. By optimizing the network structure and parameters, DRCN enhances its adaptability to small data samples and improves the generalization performance of the model. To verify this method, the DRCN model was tested in a set of bearing failure data from Case Western Reserve University (CWRU) and Jiangnan University (JNU). The results show that the proposed model achieves an accuracy of 99.13% on the CWRU dataset and 98.78% on the JNU dataset. Compared with the traditional ResNet and CapsNet methods, the DRCN proposed in this paper has higher diagnostic accuracy.
Keywords: Deep residual capsule network, equipment, fault diagnosis, production, rolling bearing.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Zhang, Y. (2026). Rolling Bearing Fault Detection Based on Deep Residual Capsule Network. Journal of Engineering, Project, and Production Management, 16(3), 2026-269.
