Multi-polynomial chaos Kriging-based adaptive moving strategy for comprehensive reliability analyses

To effectively evaluate the comprehensive reliability of structural systems, the multi-polynomial chaos Kriging-based adaptive moving strategy (AMS-MPCK) is proposed, by integrating the moving least squares (MLS) method, adaptive equilibrium optimizer (AEO) algorithm, polynomial chaos expansions, Kriging model, synchronous modeling thought and linkage sampling technique. In this strategy, the MLS is adopted to select effective samples from training samples for modeling, the developed AEO algorithm is used to obtain the optimal local compact support region radius of MLS, the polynomial chaos expansions are applied to approximate the global behavior, the Kriging model is suited to manage the local variability of output response, the synchronous modeling thought is employed to realize the synchronous construction of multiple models, and the linkage sampling technology is utilized to obtain comprehensive output response values at the same time for improving the efficiency of reliability analysis. The accuracy and efficiency advantages of the proposed AMS-MPCK are verified by the benchmark functions approximation problems, the landing gear brake system temperature, and aeroengine turbine blisk multi-failures. Besides, the developed AMS-MPCK holds excellent modeling and simulation performance by comparing with different methods. The efforts of this study provide valuable insight into the comprehensive reliability analyses of mechanical structure systems.