Temperature-Dependent Thermal Conductivity Identification by Bayesian Inference

The identification of temperature-dependent thermal conductivity in aerogel material, which is commonly used as insulation in thermal protection structures of high-speed aircraft, faces the challenge of selecting the appropriate model in engineering practice. Considering the uncertainties in the selection process of an appropriate functional model, a novel Bayesian probability method computational framework based on response data is established to improve the accuracy of thermal conductivity identification. Three implementation steps are presented: 1) the database of candidate models is established; 2) the reconstructed signals can be calculated by a heat transfer analysis model; and 3) the posterior probability of each candidate model is estimated to obtain the optimal thermal conductivity model and determine the characteristic coefficients. Numerical simulations of a theoretical one-dimensional heat transfer model and a curved thermal protection structure are performed to verify the proposed method. Then, a heating experimental investigation of the curved thermal protection structure is conducted to identify the temperature-dependent thermal conductivity of aerogel material. The results indicate that the temperature-varying thermal conductivity can be accurately identified by the proposed method, which can be applied to the heat transfer analysis and design of aerogel materials in high-speed aircraft.