Fault Diagnosis of Rotating Machine Based on Audio Signal Recognition System: An Efficient Approach

Thabit Sultan Mohammed and Mohammed Rasheed (Dhofar University, Oman); Muzhir Al-Ani (University of Human Development, Iraq); Qeethara Al-Shayea (Al-Zaytoonah University of Jordan, Jordan); Firas Alnaimi (Dhofar University, Oman)

An efficient algorithm for condition monitoring of rotating machines is proposed in this paper. Condition indicators are extracted from sound signals, and used to help undertaking the best decision about the performance state of the machine. Sound signals are recorded by microphones and processed using time-frequency domain analysis. Number of statistical features; such as mean, standard deviation, skewness, and kurtosis are extracted. The adopted statistical features are chosen because they were proven effective and simple in interpretation. Healthy, about to be faulty, and faulty performance states of the machine are considered, and audio signals are recorded for each of them. The five main steps comprising the implemented approach are data acquisition, preprocessing, feature extraction, time and frequency domain analysis, and the decision making, based on the obtained statistical measures. The experimental results indicated that an excellent recognition and distinguishing between audio signals are obtained by statistical measures in frequency domain.

Journal: International Journal of Simulation- Systems, Science and Technology- IJSSST V21

Published: Dec 30, 2020

DOI: 10.5013/IJSSST.a.21.01.08