Parameter identification of a macroscopic hardening law for metals based on crystal plasticity theory assisted by fuzzy logic

Modeling the mechanical behavior of metallic materials under loading, while accounting for both macroscopic deformation and microstructural evolution, is computationally demanding, particularly when crystal plasticity models requiring numerous hardening parameters are employed.Consequently, the reliable identification of these parameters represents a key challenge for the practical application of such models.This study investigated the application of a macroscopic phenomenological hardening function containing only four material parameters for the description of strain hardening in selected metals.An optimization-based procedure was employed to identify the hardening parameters from experimental stress-strain curves.A sensitivity analysis was additionally performed to quantify the influence of individual hardening parameters on the model response.Since the objective function may exhibit multiple local minima, a fuzzy logic-based algorithm is introduced to improve the robustness of parameter selection and identify the most representative solution under uncertainty.The proposed methodology was validated using experimental data for selected metals with different structures, demonstrating very good agreement between numerical predictions and experimental hardening curves.The results indicate that the combined optimization and fuzzy logic approach enables reliable parameter identification while significantly reducing the complexity and computational cost associated with conventional crystal plasticity hardening models.The proposed framework provides an efficient methodology for identifying the parameters of a macroscopic hardening law derived from crystal plasticity theory.Since the constitutive model was developed for subsequent application in crystal plasticity-based polycrystal modeling, the identified parameters constitute a reliable basis for their future implementation in CPFEM simulations.