A method for time‐varying reliability calculation of RC bridges considering random deterioration process

Abstract

AbstractThis paper proposes a new method based on dual neural networks to improve the efficiency of time‐varying reliability calculation of bridge structures. Based on the basic theory of structural time‐varying reliability analysis, this method uses the dyadic neural network integration method to convert the time‐varying reliability calculation of bridges from the traditional Monte Carlo simulation to the direct integration operation, which improves operation efficiency and retains high working accuracy. In addition, considering the randomness of the decay process of different bridge resistances, the variation law of time‐varying reliability of bridges in the process of uncertain resistance reduction is studied, and an explicit formula for time‐varying reliability of bridges is given and solved by using the proposed dual neural network method. Finally, the effectiveness of the proposed method is verified by an engineering arithmetic example. The result indicates that the bridge has a greater probability of failure under random conditions.