Internal Leakage Detection of Hydraulic Cylinder Based on Wavelet Analysis and Backpropagation Neural Network

摘要

Hydraulic cylinders are used as actuators of the hydraulic system of cable pendulum bar in a space launch tower. Internal leakage is a common failure mode that severely affects space launch missions. Fast and accurate identification of hydraulic cylinder leakage can ensure the safety of hydraulic cylinders and the whole system. In this paper, wavelet analysis is used to extract fault features, and a backpropagation neural network is used to establish a classifier for intelligent identification of hydraulic cylinder leakage faults. Experimental results show that the proposed method has high accuracy, providing a basis for the realization of intelligent state monitoring of launch site hydraulic systems.