Identification of Material Parameters in Enhanced Fracture Model with the Embedded Strong Discontinuity
Please login to view abstract download link
The main aim of this research is to present a Bayesian inverse identification methodology for identification of elastic and fracture parameters in quasi-brittle materials. The basis of this identification process relies on a two-dimensional solid element that is enhanced by incompatible modes and embedded strong discontinuities. Notably, this model has proven effective in accurately simulating intricate fracture processes. The identification procedure employs the Monte Carlo Markov chain method. In order to optimize efficiency and expedite the analysis, a surrogate model based on polynomial chaos expansion (PCE) for the numerical model is developed. The analysis encompasses various tests, such as the tensile fracture test and the asymmetric four-point bending test, where the identification of elastic and fracture parameters is conducted. The results of the parameter identification underscore the methodology's capability for successfully performing inverse identification of the specified parameters. Furthermore, this approach facilitates an analysis of uncertainty and sensitivity for each parameter, the results of which are also presented.
