Hydraulic Pump Fault Diagnosis Method Based on EWT Decomposition Denoising and Deep Learning on Cloud Platform

An axial piston pump fault diagnosis algorithm based on empirical wavelet transform (EWT) and one-dimensional convolutional neural network (1D-CNN) is presented. The vibration signals pressure of are taken as the analysis objects. Firstly, original decomposed by EWT, each signal component screened reconstructed according to energy characteristics. Then, time-domain features frequency-domain denoised extracted, time domain frequency fused. Finally, 1D-CNN model was deployed WISE-Platform a Service (WISE-PaaS) cloud platform realize real-time platform. Compared with ensemble mode decomposition (EEMD) complementary (CEEMD), results show that EWT has higher identification accuracy.