SA-CPSO enhanced HSMM for PHM of aircraft landing gear system

This study addresses the stochastic fault dynamics of aircraft landing gear systems under non-stationary operational conditions by proposing a Prognostics and Health Management (PHM) approach. The method enhances traditional maintenance strategies by integrating the Hidden Semi-Markov Model (HSMM) with the Self-Adaptive Chaotic Particle Swarm Optimization (SA-CPSO) algorithm. Specifically, the approach overcomes three critical limitations of conventional methods: fixed state duration assumptions in standard HSMM, susceptibility to local optima in high-dimensional parameter estimation, and delayed fault prediction during transient phases. Initially, an HSMM is developed to characterize the degradation states of the landing gear system. Subsequently, the SA-CPSO algorithm is employed to optimize the prediction model, with parameter estimation conducted via the forward-backward algorithm. Finally, the effectiveness and applicability of the optimized model are demonstrated through case studies. This PHM framework offers practical benefits by improving maintenance efficiency, enhancing safety and reliability, and reducing unscheduled faults in landing gear systems.