基于改进遗传算法的NiTi合金动态本构参数优化识别研究OA

A dynamic constitutive model of NiTi shape memory alloy constructed based on the framework of irreversible thermodynamics contains multiple undetermined constitutive parameters.To improve the efficiency and accuracy of parameter identification,the Latin Hypercube Sampling(LHS)method is adopted to sample the model parameters.Combined with the Spearman rank correlation analysis method in non-parametric statistics,the correlation between the random input sample set of constitutive parameters and the corresponding output result set of the objective function is analyzed,and the global analysis of parameter sensitivity is realized based on the Spearman rank correlation coefficient.On the basis of sensitivity analysis,an improved genetic algorithm is used for the optimal identification of dynamic constitutive parameters of NiTi alloy.The calculation results of the semi-implicit stress integration method show that the dynamic stress-strain curves of NiTi alloy obtained by numerical simulation are in good agreement with the existing experimental data,which verifies the effectiveness of the proposed optimal identification method and can be applied to the identification of constitutive parameters of NiTi alloy and other similar materials.