Multiple failure prognosis of landing gear retraction/extension system based on H_∞ filtering

Aiming at the different colored noises and multiple failure modes in engineering practices, an efficient prognostic tool based on stochastic filtering-based method is presented to predict the landing gear retraction/extension system's time to failure in this paper. Through the operational principle and failure characteristic analysis, the major failure modes and monitoring parameters are chosen for failure prognosis, and then the dependency matrix (D-matrix) which describes the relationship between these two items is built to isolate different failure modes. Afterwards, the H_∞ filter is used to track the measurement parameter and predict the time to failure with predefined confidence interval. In the end, with the D-matrix, the final failure modes for system can be isolated.