Aviation hydraulic pipeline is an important part of aircraft hydraulic system.The early weak fault vibration signal of aviation hydraulic pipeline was studied.The signal was decomposed into multiple components by using the method of complete ensemble empirical mode decomposition with adaptive noise(CEEMDAN),and the ResNet network structure was built.The obtained components were put into the deep residual network (ResNet) for training and testing.The experimental results show that the fault recognition rate of CEEMDAN-ResNet model can reach 99.78%; when the fault prediction training iteration reaches 1 200 times,the accuracy rate will reach about 99.5% and continues to be stable,which verifies the accuracy and feasibility of the CEEMDAN-ResNet model for early fault identification and prediction of aviation hydraulic pipeline.