Fault Diagnosis and Health Assessment of Landing Gear Hydraulic Retraction System based on Multi-source Information Feature Fusion

In order to solve the problems that a single signal cannot provide sufficient fault information, while the direct using of multi-sensor signals for fusion diagnosis will lead to a heavy calculation which will reduce the diagnostic efficiency, a multi-source information feature fusion method is proposed in this paper. The stacked denoising autoencoders (SDAE) is used to extract the abstract features of time-domain features of multi-source signals, and then locality preserving projection (LPP) is used to dimension reduction to complete the feature fusion. Finally, the fused low-dimensional features act as inputs to the support vector machine (SVM) to realize the failure detection and fault location of typical fault modes of the landing gear hydraulic retraction system. The inhibitory effect of the closed-loop system on the incipient fault is discussed as well. Moreover, a health assessment method is presented considering the gradual degradation of leakage fault of the actuator. The results show that the proposed method is more accurate and reliable than any single signal result. The model of health assessment can give the internal leakage severity of the actuator. The significance of this paper is to provide a feasible idea of the fault diagnosis and health assessment of the landing gear hydraulic retraction system.