Fault detection of gearbox from vibration signals using time-frequency domain averaging

The vibration signal of a gearbox carries the signature of the fault in the gears, and early fault detection of the gearbox is possible by analyzing the vibration signal using different signal processing techniques. Time domain average can extract the periodic waveforms of a noisy vibration signal, whereas Wavelet transformation is able to characterize the local features of the signal in different scales. This paper proposes a new technique, Time-Frequency Domain Average, which combines the time domain average and wavelet transformation together to extract the periodic waveforms at different scales from noisy vibration signals. The technique efficiently cleans up noise and detects both local and distributed faults simultaneously. A pilot plant case study has been presented to demonstrate the efficacy of the proposed technique.