Adaptive inverse analysis (AIA) applied and verified on various fiber reinforced concrete composites

During the past decades several inverse approaches have been developed to identify the stress-crack opening (σ- w) by means of indirect test methods, such as the notched three point bending-, wedge splitting-, and round panel testing. The aim is to establish reliable constitutive models for the tensile behavior of fiber reinforced concrete materials, suitable for structural design. Within this context, the adaptive inverse analysis (AIA) was recently developed to facilitate a fully general and automatized inverse analysis scheme, which is applicable in conjunction with analytical or finite element simulation of the experimental response. This paper presents a new formulation of the adaptive refinement criterion of the AIA method. The paper demonstrates that the refinement criterion of the nonlinear least square curve fitting process, is significantly improved by coupling the model error to the crack mouth opening and the crack opening displacement relationship (w_cmod- w_cod). This enables an adaptive refinement of the σ- w model in the line segment with maximum model error, which entails significant improvement of the numerical efficiency of the AIA method without any loss of robustness. The improved method is applied on various fiber reinforced concrete composites and the results are benchmarked with the inverse analysis method suggested by the Japanese Concrete Institute (Method of test for fracture energy of concrete by use of notched beam, Japanese Concrete Institute Standard, Tokyo, 2003) and recently adopted in ISO 19044 (Test methods for fibre-reinforced cementitious composites—load-displacement curve using notched specimen, 2015). The benchmarking demonstrates that the AIA method, in contradiction to the JCI/ISO method, facilitates direct determination of the tensile strength and operational multi-linear σ- w models.