Fast Fourier Transform and Discrete Wavelet Transform for Fault Diagnosis of Gear

Ahmed, Rounaq and Pai, Srinivasa P and Sriram, N S and Vijay, G S (2013) Fast Fourier Transform and Discrete Wavelet Transform for Fault Diagnosis of Gear. In: International Conference on Emerging Trends in Engineering, 15 and 16 May 2013, NITTE Karnataka.

PDF paper 13.pdf - Published Version Restricted to Registered users only Download (4MB) | Request a copy

Abstract

Vibration analysis plays an important role in condition monitoring and fault diagnosis of gears. Many signal analysis methods are used to extract the information from the data. In this paper the Fast Fourier Transform (FFT) and Discrete Wavelet Transform (DWT) analysis of the raw vibration signals is carried out. It is found that there is amplitude variation at Gear Mesh Frequency (GMF) and side bands of the GMF with respect to the defect type. The energy and kurtosis values of the DWT approximation coefficients can be used for fault diagnosis of Gear. Further the approximation coefficients obtained from DWT are subjected to FFT analysis. The FFT analysis of the DWT approximation coefficients gives better results when compared to direct FFT analysis of raw signals for fault identification. The study has been conducted on the labeled data of PHM’09 challenge. The data included the recordings of vibration in spur gear box with seeded faults.

Actions (login required)

View Item