A Fault Diagnosis Study of Aero-engine Hydraulic Piping Clamp Based on Deep Learning

The characteristics of the fixed clamp in the hydraulic piping of aero-engine are not obvious, the accuracy of fault identification is low, the fault analysis methods based on human and expert knowledge is limited. In order to address the above problems, a method was proposed, based on which, the optimized VMD was combined with Deep Convolutional Neural Network to create an intelligent fault diagnosis method for VMD and DCNN clamps. Firstly, the signal was decomposed by the optimized VMD, and then DCNN was adopted for fault pattern identification, finally this method was applied in example. The results showed that this method not only could decompose the signal accurately and effectively, but also identify the clamp fault type with 98.2% accuracy.