Rolling bearing fault diagnosis based on WOA-VMD-YOLOv8
Zihui Ding
College of Information Engineering, Jiangsu Maritime Institute
Chenyang Wang
College of Information Engineering, Jiangsu Maritime Institute
Di Zhang
College of Information Engineering, Jiangsu Maritime Institute
Xin Hu
College of Information Engineering, Jiangsu Maritime Institute
Jin Wang
College of Information Engineering, Jiangsu Maritime Institute
Sheng Chen
College of Information Engineering, Jiangsu Maritime Institute
DOI: https://doi.org/10.59429/esta.v12i1.9655
Keywords: Fault diagnosis; Whale optimization algorithm; Variational mode decomposition; YOLOv8
Abstract
Rolling bearings are often in a state of high-speed operation, making them highly susceptible to failure. To address this issue, the WOA-VMD-YOLOv8 fault diagnosis model was designed by combining the advantages of Whale optimization algorithm (WOA), variational mode decomposition (VMD) and YOLOv8. The bearing vibration signals collected by Case Western Reserve University were used as experimental data. WOA was applied to optimize the VMD parameters, and the time-frequency diagrams of the WOA-VMD output signals were used as the dataset for YOLOv8. The experimental results show that, compared to the Wavelet Transform(WT)-YOLOv8, the WOA-VMD-YOLOv8 achieves better predictive performance for bearing faults, with Recall rate of 99.7%, mAP@0.5 of 99.5%, and mAP@0.5-0.95 of 99.4%. Therefore, the WOA-VMDYOLOv8 demonstrates practical value in bearing fault detection technology.
References
[1] X. Zhang, Q. Miao, Z. W. Liu, et al. An Adaptive Stochastic Resonance Method Based on Grey Wolf Optimizer Algorithm and Its Application to Machinery Fault Diagnosis[J], ISA transactions, 2017, 71: 206-214.
[2] G. Wang, J. W. Xiang. Remain useful life prediction of rolling bearings based on exponential model optimized by gradient method[J], Measurement, 2021, 176:109161.
[3] M. M. Manjurul Islam, Alexander E. Prosvirin, Jong-Myon Kim. Data-driven prognostic scheme for rolling-element bearings using a new health index and variants of least-square support vector machines[J], Mechanical systems and signal processing, 2021, 160:107853.
[4] L. B. Cosme, W. M. Caminhas, M. F. S. V. D’Angelo, et al. A Novel Fault-Prognostic Approach Based on Interacting Multiple Model Filters and Fuzzy Systems[J], IEEE Transactions on Industrial Electronics, 2019, 66(1):519-528.