Physics

Finite element method

Brittleness

Nonlinear system

Cementitious

Structural engineering

Bending

Engineering

Artificial intelligence

Composite material

Materials science

Computer science

Probabilistic logic

Cement

Quantum mechanics

Fracture (geology)

Fracture mechanics

Advanced materials research

The scatter of experimental results using specimens made of quasi-brittle materials, such as concrete, fibre-reinforced concrete, ultra high performance concrete etc., can be due to their heterogeneity rather high. An assessment of fracture-mechanical parameters is then difficult and problematic. To remain at deterministic level is therefore unrealistic and without virtual statistical approach, simulation and probabilistic result assessment the consequent practical design of quasi-brittle material-based structures can be risky. A key parameter of nonlinear fracture mechanics modeling is fracture energy of concrete. Numerical simulation of concrete failure and fracture phenomena in concrete as well as other cementitious materials became a field of an intensive research in the recent years. With respect to accuracy and efficiency of corresponding numerical models some few still open questions have to be focused. How the heterogeneity of cementitious materials can be taken into consideration in the most realistic way using commercially available finite element programs A sophisticated option to get the parameters of the computational model indirectly is based on combination of fracture test with inverse analysis. This paper describes a methodology to get such parameters using experimental data from three-point bending tests used in inverse analysis based on combination of artificial neural networks and stochastic analysis.

Fracture-Mechanical Parameters for Modeling of Quasi-Brittle Materials and Structures