Identification of Chaboche-Lemaitre combined isotropic-kinematic hardening model parameters assisted by the fuzzy logic analysis

A very good knowledge of material properties is required in the analysis of severe plastic deformation problems in which the classical material processing methods are accelerated by the application of the additional cyclic load. A general fuzzy logic-based approach is proposed for the analysis of experimental and numerical data in this paper. As an application of the fuzzy analysis, the calibration of Chaboche-Lemaitre model hardening parameters of PA6 aluminum is considered here. The experimental data obtained in a symmetrical strain-controlled cyclic tension-compression test were used to estimate the material's hardening parameters. The numerically generated curves were compared to the experimental ones. For better fitting of numerical and experimental results, the optimization approach using the least-square method was applied. Unfortunately, commonly accepted calibration methods can provide various sets of hardening parameters. In order to choose the most reliable set, the fuzzy analysis was used. Primarily selected values of hardening parameters were assumed to be fuzzy input parameters. The error of the hysteresis loop approximation for each set was used to compute its membership function. The discrete value of this error was obtained in the defuzzification step. The correct selections of hardening parameters were verified in ratcheting and mean stress relaxation tests. The application of the fuzzy analysis has improved the convergence between experimental and numerical stress-strain curves. The fuzzy logic allows analyzing the variation of elastic-plastic material response when some imprecisions or uncertainties of input parameters are taken into consideration.

Automated summary

A fuzzy logic-based approach is proposed to analyze severe plastic deformation problems, focusing on the calibration of hardening parameters in material models. Experimental data from tension-compression tests were used to estimate the material's hardening parameters, and an optimization method was applied to improve the fitting of numerical and experimental results. The fuzzy analysis enhanced the convergence between experimental and numerical stress-strain curves by considering imprecisions and uncertainties in input parameters.