Inverse Problem of Probability of Concrete Biaxial Strength Test

Abstract:

On basis of the probability density evolution theory and stochastic damage mechanics of concrete， a class of stochastic modeling principle is established from the perspective of probability density. In the context of concrete biaxial strength tests， a method is proposed to solve the inverse problem of probability based on the genetic algorithm. Utilizing the test data and the proposed method， the statistical parameters of fracture strain in the elastoplastic stochastic damage model for concrete are identified， and the basic parameters of the model are obtained. It is demonstrated that the identified parameters can faithfully capture the randomness of concrete strength at both probability density and statistical moment levels. The study of this paper aims to provide a new road for solving the inverse problem of probability.