Identifying material parameters in crystal plasticity by Bayesian optimization

Abstract In this work, we advocate using Bayesian techniques for inversely identifying material parameters multiscale crystal plasticity models. Multiscale approaches modeling polycrystalline materials may significantly reduce the effort necessary characterizing such models experimentally, in particular when a large number of cycles is considered, as typical fatigue applications. Even appropriate microstructures and microscopic are identified, calibrating individual model to some experimental data industrial use, task formidable even single simulation run time consuming (although less expensive than corresponding experiment). For solving problem, investigate Gaussian process based optimization, which iteratively builds up improves surrogate objective function, at same accounting uncertainties encountered during optimization process. We describe approach detail, high-strength steel an application. demonstrate that proposed method upon comparable on evolutionary algorithm performing derivative-free methods.