Research On Thermophysical Parameter Identification Of Pyrolysis Process Of Thermal Protection Materials

In the design of thermal protection system,the ablative thermal protection material is more and more widely used in the re-entry capsule and interplanetary exploration due to its high thermal protection efficiency and low density.The thermo physical parameters of ablative materials are very important for the internal heat conduction and pyrolysis of materials,which is hard to accurately acquire because of pyrolysis reaction happening when heated.In this paper,the thermal property parameters in the pyrolysis process of materials are identified by using the inverse heat transfer problem method.thermal protection material is divided into virgin material,pyrolyzing material and carbonized material according to pyrolizing and carbonizing temperature.As to virgin material parameter identification problem,this paper puts forward method of LM segment identification method,improving the global convergence of LM method;for simultaneously identification of thermal conductivity and specific heat capacity,and a novel iterative format is put forward for the identification of temperature dependent thermal conductivity and specific heat capacity of carbonized material.Through a numerical example,the identification precision is verified.By superposition of random errors in the temperature data,the stability of the parameter identification method for random error is verified.The parameter identification experiment is carried out with a laser heater,verifying the effectiveness of the algorithm.Through the analysis of thermo gravimetric experiment,the pyrolysis kinetic parameters of the material are obtained.Material ablation process simulation model is established and used to calculate the corresponding temperature response of the training samples of parameters,which is obtained by the maximin Latin hypercube sampling method.Through the principal component analysis the dimension of temperature response is reduced,resulting in the training sample set,containing the samples of parameters and principal components of corresponding temperature response.The Gaussian process model is used to construct surrogate models.The combination kernel function of square exponent and Matern 3/2 is picked out by comparing the prediction performance.The influence of the size training samples on the prediction accuracy of the surrogate model is investigated to obtain the training size is 25 times the number of parameters.The calculation efficiency of the surrogate model is apparently improved.The ME-AMH sampling method combining the maximum entropy principle and the adaptive MH method is proposed to sample from the parameter posterior distribution,and the strategy of iterating ME-AMH method to cyclically modify the surrogate model for material parameter identification is proposed.Taking the temperature response calculated by the simulation model as the temperature measurement value and using the Gaussian process model,parameters of the pyrolysis process are identified,indicating that the training samples of only 10 times of parameters are needed,verifying the validity efficiency and stability of the iterative ME-AMH method for parameter identification.The non-oxide wind tunnel experiment are designed and carried out for light carbon phenolic thermal protection material,obtaining the temperature response history of the virgin and carbonized material The temperature dependent thermal parameters of the virgin carbonized material are identified using the proposed method The repeatability between different models is good,which is basically consistent with the identification results of the thermostatic heating platform for the virgin material.The thermal conductivity of the carbonized material increase from 0.1 W/(m·K)to 0.7W/(m·K),and specific heat capacity increase from 700 J/(kg·K)to 2970 J/(kg·K)and then decrease slowly to 2610 J/(kg·K),showing good repeatability between different models.Utilizing the temperature information of light carbon phenolic thermal protection material in the on-oxide wind tunnel experiment,combined with the identification results of the virgin and carbonized material,the material ablation surrogate model is established.The surface absorption coefficient and temperature dependent thermal conductivity of the material in pyrolysis process and their confidence intervals are identified using the experimental temperature data,the ablation surrogate model and the iterative ME-AMH method proposed in this paper.The results of pyrolysis process parameter identification are compared with those of virgin materials and carbonized materials to prove the reliability of pyrolysis process parameter identification results.Finally,the temperature response under step heat flux condition is predicted with the pyrolysis process parameter identification results,and the prediction is compared with those of wind tunnel experiment,improving the prodiction precision compared to the prediction result using parameters of virgin and carbonized material.,and proving the effectiveness and reliabilityof parameter identification results,and then the material responses under heat flux of service conditions can be predicted to guide the design of thermal protection system.