Fault diagnosis of fracturing truck based on SSST and DRN

XU Xu1, CHEN Zhigang1,2, DU Xiaolei1, ZHANG Nan1, ZHONG Xinrong3

(1.School of MechanicalElectrical and Vehicle Engineering, Beijing University of Civil Engineering and

Architecture, Beijing 100044, China; 2.Construction Safety Monitoring Engineering Technology

Research Center of Beijing, Beijing 100037, China; 3.Changqing Downhole Technology Company,

CNPC Chuanqing Engineering Company Limited, Xi’an 710021, China)

Abstract: Aiming at the problem that it is difficult to accurately extract and identify the hydraulic end fault of fracturing truck under complicated working conditions and high load environment, combining the advantages of deep residual network, a fault diagnosis method based on synchrosqueezed S transform（SSST） and deep residual network (DRN) was proposed. Firstly, based on the superior timefrequency decomposition characteristics of SSST, the vibration signals collected by the power end of 2000 fracturing truck was decomposed and transformed to obtain timefrequency images. Then, the timefrequency image was grayed and normalized, the grayscale was reduced to an appropriate size, and the compressed timefrequency image was used as the input of the DRN. Finally, a classification recognition model based on SSST feature extraction and DRN was established，and tested to realize the fault identification of the power end of the fracturing truck.The results indicate that the method avoids the complex process of artificial feature extraction，and can effectively improve the accuracy of fault state identification of power end of fracturing truck under strong background noise.