Development of a Synchronous Filter for Time Domain and Rotation Domain Averaging

The interaction of the various components in rotating machinery like gearboxes may generate excitation forces at various frequencies. These frequencies may sometimes overlap with the frequencies of the forces generated by other components in the system. Conventional vibration spectrum analysis cannot attenuate noise and spectral frequency band overlapping, which in many applications mask the change in the structural response due to the deterioration in a certain component of the machine. This problem is overcome by time domain averaging. In time domain averaging, the vibration signal is sampled at a frequency that is synchronised with the rotation of the gear of interest and the samples obtained for each singular position of the gear are then ensemble-averaged. The time domain averaging or synchronous averaging procedure requires an enormous amount of vibration data, making it very difficult to develop online condition monitoring systems for gearboxes since data acquisition and analysis cannot be done simultaneously. This paper presents the use of artificial neural networks in the development of a time domain averaging model for addressing this problem. In addition a comparison is made between different neural network formulations in this application. The results indicate that artificial neural networks are suitable for use in domain averaging.