On the use of probabilistic and non-probabilistic super parametric hybrid models for time-variant reliability analysis

Due to having a large number of uncertain variables originating from various sources in time-variant reliability problem, the methods for taking account of these uncertainties are of high significance in engineering. In this study, a novel hybrid time-variant reliability model is established based on a probabilistic model and a super parametric convex model. The probabilistic and non-probabilistic parameters can be addressed simultaneously using this model. The stochastic processes are discretized using the expansion optimal linear estimation method to capture the correlation among different time nodes. In addition, the hybrid reliability iteration method is presented to obtain the upper and lower bounds of failure probability simultaneously. Subsequently, a new relaxed reliability iteration method is proposed to solve the hybrid time-variant reliability model. The relaxed method is adopted to ensure the robustness and efficiency of the proposed method. Three examples are discussed to demonstrate the validity and usage of the presented methodology.