Acoustic emission detection in concrete specimens: Experimental analysis and lattice model simulations

In civil engineering, a quantitative evaluation of damage in materials subjected to stress or strain states is of great importance due to the critical character of these phenomena, which may suddenly give rise to catastrophic failure. From an experimental point of view, an effective damage assessment criterion is provided by the statistical analysis of the amplitude distribution of the acoustic emission signals generated by growing microcracks. A classical way to work out the amplitude of acoustic emission signals distribution is the Gutenberg-Richter law, characterized by the b-value parameter, which systematically decreases with damage growth. The damage process is also characterized by a progressive localization that can be modeled through the fractal dimension 2 b = D of the damaged domain. In the framework of continuum damage mechanics, the progressive deterioration of the material that causes formation of macro-cracks is described by means of phenomenological damage variables usually introduced in classical constitutive relationships. Nevertheless, taking into account discrete damage mechanics, lattice models are particularly suitable to reproduce the generation of acoustic emission events, arising from the materials, during the different stages of damage growth. These models are also fundamental for the application of advanced statistical methods and non-standard mathematical methods, e.g. fractal theory. Starting from these considerations, in this work a b-value analysis was conducted in laboratory on two concrete specimens loaded up to failure. One was a prismatic specimen subjected to uniaxial compressive loading and the other was a pre-cracked beam subjected to a three-point bending test. The truss-like discrete element method was used to perform numerical simulations of the testing processes. The test results and the results of the numerical analyses, in terms of load vs. time diagram and acoustic emission data, as determined through b-value and signal frequency variations, are compared and are seen to be in good agreement.