Service Life Prediction Model of Reinforced Concrete Structure based on Evolutionary Neural Network

Abstract

When the traditional model predicts the service life of a concrete structure, the number of iterations of its stress and strain parameter points is less, which leads to errors in the calculation value of the concrete structure's bearing capacity. To this end, a prediction model for the service life of reinforced concrete structures based on evolutionary neural networks is constructed. Use different material properties of reinforced concrete structure to numerically simulate multiple mechanical properties, use evolutionary neural network, evolve mechanical parameter points, use concrete buckling strength and buckling bearing capacity as the main parameters of the model, obtain the parameter limit load value, and use the full probability Distribution method, the decay time of the concrete structure is obtained, which is the remaining service life. Comparing this model with two traditional models, the results show that this model reduces the calculation error of the structural bearing capacity, the model input parameters are more accurate, and the reliability of the life prediction value of the reinforced concrete structure is improved.