Diagnostics of rotating machines at constant and variable speeds

The main objective of this book is to develop a method for diagnosing gear defects in stationary and variable conditions, based on the complementarity between objective and subjective approaches. The objective approach is based on vibration analysis, which detects anomalies from signals, regardless of speed variations. The subjective approach relies on sound perception to identify abnormal sounds associated with mechanical defects. After a comparative study of these two methods in noisy environments, an optimised hybrid method was proposed, combining ICEEMDAN, wavelet denoising and order analysis. The results show an increased ability to detect gear defects in non-stationary conditions. Perceptual tests carried out at ENSIM (Le Mans, France) experimentally validated the proposed approach in a realistic context, highlighting the value of combining vibration and auditory analyses for reliable diagnosis of rotating machines.