Answers are grounded in this paper and cite their sources.
High-speed rail grinding train operates at high speed. If grinding wheel faults such as cracking and chipping exist, they may seriously compromise the grinding quality and safety. In the past, fault identification accuracy of real-time online monitoring of grinding wheels based on local mean decomposition (LMD) of vibration signal was limited by the end effects of traditional LMD. To identify grinding wheel faults timely and accurately, an improved LMD method for grinding wheel was proposed. First, grinding experiments using grinding wheels with typical faults, including single crack, two cracks, and single chipping, were conducted to obtain the vibration characteristics of damaged grinding wheels. Then, a signal extension method based on waveform matching was established to solve the end effects problem of traditional LMD. Finally, fault diagnosis of high-speed rail grinding wheels was realized based on the improved LMD and support vector machine (SVM). The results show that the improved fault diagnosis method solved the fault underreporting problem that exists in traditional method. The comprehensive accuracy of recognizing the three typical faults increases from 84.62% to 94.87%. Specifically, the recognition accuracy for the most commonly seen fault type of single crack increases by 15.38%, which significantly alleviates the low recognition accuracy problem of traditional methods and ensures the safety and quality of high-speed grinding.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: