Hybrid identification method of material parameters based on genetic algorithm and improved homotopy algorithm

This paper proposes a hybrid identification method of material parameters based on genetic algorithm and improved homotopy algorithm. In this method, genetic algorithm is first used to perform a preliminary identification of the model parameters. Then, the results of the preliminary identification are used as the initial values for more precise parameter identification using the improved homotopy algorithm. Based on the Euler prediction-Newton correction homotopy algorithm, a curve prediction-Newton correction homotopy algorithm is proposed to improve the calculation accuracy and efficiency. The effectiveness and accuracy of the hybrid identification method are verified by numerical examples. The hybrid identification method is applied successfully to determine material characteristic parameters of FGH97. The results indicate that based on experimental data and numerical simulations, the hybrid identification method can rapidly obtain effective and reliable material parameters. This method reduces the calculation amount, avoids dependence on the selection of the initial population, and has a high calculation accuracy and efficiency for the inverse problem of parameter identification. It provides an effective method to accurately identify material parameters and can also be applied to parameter identification of other materials.