Dynamic Modeling of the Gear Transmission for Detecting the Separate and Simultaneous Tooth Fracture Using the Statistical Process Control Technique

Dynamic models of transmission systems have long been used to study the behavior of vibration response and its impact on gear fault detection. This work proposes a method for detecting separated and simultaneous fractures caused by a variation in gear mesh stiffness (GMS) for a dynamic model of a spiral bevel gear system (SBGS) using statistical process control schemes (SPC). Two types of SPC, known as the X-bar chart and an exponentially weighted moving average (EWMA) chart, have been applied. Initially, noise was added to simulate dynamic modeling signal conditions in practice by using different levels of noise variance. Second, the control limits of both X-bar and EWMA control charts are designed using the relative wavelet energy (RWE) computed from the frequency subbands of discrete wavelet transform analysis. Third, the performance of control schemes for separated and simultaneous fracture damage detection is evaluated based on the RWE features of simulated vibration signals under defective conditions. The results of the proposed method indicated that the EWMA chart outperformed the X-bar chart in detecting separated and simultaneous fracture faults at an early stage.