Matthew Aladesaye

Citation

Thesis Title
Application of Predictive Maintenance to Industry Including Cepstrum Analysis of a Gearbox

Read article at Massey Research Online:

Mr Aladesaye investigated current maintenance practice in several New Zealand industries and the improvements that could be obtained by the use of predictive maintenance techniques. He undertook a series of case studies within New Zealand industries situated in Auckland. Results showed a well defined preventive maintenance schedule embedded into a systems, applications and products programme to monitor routine maintenance. He further investigated the current predictive maintenance that uses a Fast Fourier Transform algorithm and discovered its limitations in diagnosing gearbox faults. He finally looked at an inverse Fast Fourier Transform, called cepstrum, which allows the deconvolution of vibration signals from separate sources. Experimental results showed that the sidebands present in complicated gearbox signals were separated by the cepstrum technique from the forcing frequency. He concluded the thesis by showing how the gearbox transmission path and gearmesh effects were separated.

Supervisors
Dr Huub Bakker
Professor Johan Potgieter
Professor B Randall