Multidisciplinary systems will remain in transient states when time-dependent interactions are present among the coupling variables. This brings significant challenges to time-dependent multidisciplinary system reliability analysis. This paper develops an adaptive surrogate modeling approach (ASMA) for multidisciplinary system reliability analysis under time-dependent uncertainty. The proposed framework consists of three modules, namely initialization, uncertainty propagation, and three-level global sensitivity analysis (GSA). The first two modules check the quality of the surrogate models and determine when and where we should refine the surrogate models. Approaches are then proposed to estimate the potential error of the failure probability estimate and determine the location of the new training point. In the third module (i.e. three-level GSA), a method is developed to decide which surrogate model to refine, through GSA at three different levels. These three modules are integrated together systematically and enable us to adaptively allocate the computational resources to refine different surrogate models in the system and thus achieve high accuracy and efficiency in time-dependent multidisciplinary system reliability analysis. Results of two numerical examples demonstrate the effectiveness of the proposed framework.
Accelerated Monte Carlo system reliability analysis through machine-learning-based surrogate models of network connectivity
