Abstract
To describe the material behaviour at high strain rates dynamic experimental tests are necessary, and appropriate constitutive models are to be calibrated accordingly. A way to achieve this is through an inverse procedure, based on the minimization of an error function calculated as the difference between experimental and numerical data coming from Finite Element analysis. This approach, widely used in the literature, has a heavy computational cost associated with the minimization process that requires, for each variation of the material model parameters, the execution of FE calculations. In this work, a faster but yet effective calibration procedure is studied Experimental tests were performed on an aluminium alloy AA6061-T6, by means of a direct tension-compression Split Hopkinson bar. A fast camera with a resolution of 192 × 128 pixels and capable of a sample rate of 100,000 fps captured images of the deformation process undergone by the samples during the tests. The profile of the sample obtained after the image binarization and processing, was postprocessed to derive the deformation history; afterwards it was possible to calculate the true stress and strain, and carry out the inverse calibration by analytical computations. The results of this method were compared with the ones coming from the Finite Element approach