Detection of faults in machines using power signatures and assessment of faults on energy consumption

Abstract

Technological advancements saw the increasing reliance on machinery to perform complex and sophisticated functions. It is crucial to prevent failures from occurring to machines so as to reduce costs and lost earnings. A considerable amount of research has been done in the field of condition monitoring of machines for early detection of faults. In this project, an algorithm was proposed for diagnosing common faults of machines using the vibration signature of the machine. The proposed algorithm used the Fast Fourier Transform (FFT) and a proposed method of Feature Extraction and Fault Diagnosis. The algorithm was tested on data obtained from the Singapore Institute of Manufacturing Technology (SIMTech). The results of diagnosing a machine with unbalanced fault and a machine with bearing outer raceway fault with the proposed algorithm has a success rate of at least 99% and 69% respectively.