Application of Multi-sensor Information Fusion in the Fault Diagnosis of Hydraulic System

Aiming at the problem of incomplete information and uncertainties in the diagnosis of complex system by using single parameter, a new method of multi-sensor information fusion fault diagnosis based on BP neural network and D-S evidence theory is proposed. In order to simplify the structure of BP neural network, two parallel BP neural networks are used to diagnose the fault data at first; and then, using the evidence theory to fuse the local diagnostic results, the accurate inference of the inaccurate information is realized, and the accurate diagnosis result is obtained. The method is applied to the fault diagnosis of the hydraulic driven servo system(HDSS) in a certain type of rocket launcher, which realizes the fault location and diagnosis of the main components of the hydraulic driven servo system, and effectively improves the reliability of the system.