بهبود تشخیص و ارزیابی شدت عیوب سیستم‌های دنده‌ای: مطالعه مقایسه‌ای تکنیک‌های چگالی انرژی لحظه‌ای و تجزیه سیگنال

Early detection of gear faults plays a crucial role in enhancing safety and implementing pre-dictive maintenance strategies. In this study, a novel method based on the analysis of instan-taneous energy density of vibration signals is proposed for identifying and assessing the se-verity of faults. To this end, energy-based features are extracted using techniques such as Continuous Wavelet Transform (CWT), Empirical Mode Decomposition (EMD), and Varia-tional Mode Decomposition (VMD). Experiments were conducted on a test rig with root crack faults at varying severities. The results demonstrate that the instantaneous energy den-sity exhibits significant peaks proportional to fault severity during the engagement of the faulty tooth, confirming the superiority of this approach over traditional methods. Among the techniques, the combined VMD and CWT method showed superior performance with high sensitivity in early fault detection. These findings highlight the high potential of the proposed method for implementation in condition monitoring systems across industries.