Fault Diagnosis of Fracturing Truck Based on Variational Mode Decomposition and Deep Belief Network

Abstract

Due to the problem that it is difficult to accurately extract and identify the hydraulic end fault of 2000 fracturing truck under complicated working conditions and high load environment, a variational mode decomposition (VMD) with deep belief network (DBN) is presented. Firstly, the variational mode decomposition is used to decompose the vibration signal collected by the hydraulic end of the fracturing vehicle into several stable intrinsic mode function (IMF) and obtain the spectrum of the reconstructed signal, which is the input of the deep belief network. Then, the deep belief network fault identification model was constructed by using the back-propagation algorithm and, the pre-training and feature learning of input spectrum are carried out, the DBN-based fault feature adaptive analysis and fault state intelligent identification is completed, realizing the fault diagnosis of the hydraulic end of the fracturing truck. The results show that the adaptive characteristic of the DBN method can effectively improve the accuracy of fault state recognition.