Condition Monitoring of Gear Box using Acoustic Signals
Condition monitoring is a process to monitor the respective parameters of specified machinery to identify any occurrence of significant changes over a period of time. Condition monitoring is a predictive approach that predicts the condition of specified machinery by using sensor data that measures different parameters. Gear system is one of the power transmission systems that is used to transfer motion and torque between machine components. Gear is imposed to various kinds of loads and with different kinds of speeds that lead to wear out of tooth profile and/or other anomalies. Condition monitoring of gears plays a very important role in the detection of gear abnormalities which helps prevent catastrophic failure before the fault progresses. In this research work, condition of the gear is analysed using signal processing techniques, based on Sound signal. The signals are collected by microphone along with the data acquisition system during the operation. Signal processing techniques such as Fast Fourier Transform (FFT) and Continuous Wavelet Transform (CWT) are used to monitor the condition of gears. Amongst all frequencies of the gear, 800 Hz is the dominant frequency which recognizes the different conditions in spectrum plot whereas in CWT plots, the variations in 3D plots for different conditions of gears can be seen. Hence these techniques can be recommended for condition monitoring of gears.