Study on Fault Diagnosis Method for Gear Pump Based on Support Vector Domain Description and Distance Measure

PDF(1108 KB)

Journal of Vibration and Shock ›› 2013, Vol. 32 ›› Issue (11) : 62-65.

论文

Study on Fault Diagnosis Method for Gear Pump Based on Support Vector Domain Description and Distance Measure

Wang Tao Li Aihua Wang Xuping Cai Yanping

Author information +

History +

Abstract

A fault diagnosis method for gear pump based on support vector domain description and distance measure is put forward. Firstly, vibration signals under various conditions of the gear pump are decomposed by wavelet packet to extract energy percentage of each frequency band as feature vectors. Next SVDD hypersphere model is trained only using feature vectors under normal condition, and abnormal condition of gear pump is detected through the definition of absolute distance measure. Finally SVDD hypersphere models are trained independently by using feature vectors under each conditions, and various fault conditions of gear pump are recognized accurately through the definition of relative distance measure. Experimental results show that the data dimension is reduced and the diagnosis information is concentrated effectively by adopting feature extraction with wavelet packet technology. SVDD fault diagnosis methods based on absolute distance and relative distance measures not only can detect abnormal condition, but also can distinguish various fault conditions, achieving the purpose of condition monitoring and fault classification.

Key words

Wavelet packet decomposition / Support vector domain description / Distance measure / Gear pump / Fault diagnosis

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Wang Tao Li Aihua Wang Xuping Cai Yanping. Study on Fault Diagnosis Method for Gear Pump Based on Support Vector Domain Description and Distance Measure[J]. Journal of Vibration and Shock, 2013, 32(11): 62-65