Abstract：In order to estimate and calibrate the thermo physical parameters based on the temperature measured from the material, the processes combined with Levenberg-Marquardt (L-M) optimization algorithm, one-dimensional nonlinear and unsteady heat conduction equations for identification of temperature-dependent thermophysical parameters and confidence interval estimation of parameters are proposed in this paper. Through two numerical examples to evaluate the method in this paper, the results show that：1) The accuracy of thermophysical parameters identification is affected by fixed parameters assignment. When the fixed parameters are assigned true values, the identification results of other parameters are unbiased, conversely the identification results are biased. 2) The uncertainty of the identification results is positively correlated with the noise level of identification data. 3) When the thermal conductivities and specific heat capacities are identified at the same time, the specific heat capacities identification accuracy is higher than the thermal conductivities. 4) Both F-test method and asymptotic standard error can estimate the confidence interval of the identification parameters and reflect their degree of uncertainty. The F-test method can also reflect the degree of nonlinearity of the uncertainty. From the results of two examples, it can be seen that the identification method for temperature-dependent thermophysical parameters proposed in this paper is feasible and the accuracy meets the requirements of engineering application.
窦怡彬, 陈昂, 陆云超, 刘陆广, 李宗阳. 基于Levenberg-Marquardt算法的材料温度相关热物性参数辨识[J]. 空天防御, 2022, 5(2): 17-26.
DOU Yibin, CHEN Ang, LU Yunchao, LIU Luguang, LI Zongyang. Identification of Temperature-Dependent Thermophysical Parameters Based on Levenberg-Marquardt Algorithm. Air & Space Defense, 2022, 5(2): 17-26.
