Feature extraction of gear fault signal based on Sobel operator and WHT

Article type: Research Article

Authors: Cai, Jian-Hua | Hu, Wei-Wen

Affiliations: Department of Physics and Electronics, Hunan University of Arts and Science, Changde, Hunan, China

Note: [] Corresponding author: Jian-Hua Cai, Department of Physics and Electronics, Hunan University of Arts and Science, Changde 415000, Hunan, China. Tel.: +86 073 6718 6121; Fax: +86 073 6718 6121; E-mail: [email protected]

Abstract: Taking Wigner-Ville distribution of gear fault signal as a picture, Sobel operator was applied for edge detection of picture and then Hough transform was used to extract signal feature. Some simulated and measured signals have been processed to demonstrate the effectiveness of new method, which was compared with traditional Wigner-Hough transform and SPWD-Hough transform. The results show that the proposed method can suppress cross term which is produced from using Wigner-Ville distribution to analyze multi-component signal, especially under the condition of low signal to noise ratio. The improved Wigner-Hough transform can effectively suppress the influence of noise and has a good real-time performance because its algorithm is fast. The proposed method provides an effective method to determine the state of gear accurately.

Keywords: Wigner-Ville distribution, hough transform, Sobel operator, feature extraction of gears

DOI: 10.3233/SAV-130767

Journal: Shock and Vibration, vol. 20, no. 3, pp. 551-559, 2013