Heat Transfer Parameters And Heat Flux Identification Of The Composite Sandwich Structure

The composite sandwich structure is a kind of lightweight structure with excellent heatshielding performance and has been widely used in the thermal protection system of high-speed vehicles.Determining the severe heat flux of high-speed vehicles in the service environment is important for the designing of composite sandwich structures.Due to the limitations of sensor installation conditions,it is costly and difficult to directly measure the heat flux of high-speed vehicles under service loads.Therefore,it is an effective means to obtain the heat flux by solving the inverse heat transfer problem.For solving the inverse heat transfer problem,temperature sensors are placed on the inner side of vehicle to measure the temperature of the inner wall,which is used to identify the heat flux of the outer wall.In this paper,the composite sandwich structure is taken as the research object.The corresponding heat transfer model is established and updated based on the heat transfer experiment.The Conjugate Gradient Method(CGM)and neural network method are carried out to identify the time-varying heat flux through simulation research and experimental verification.The main work of this paper is as follows:First,the heat transfer model modification method of composite sandwich structure based on material parameter identification is proposed.The two-dimensional transient heat transfer model of the structure is established to identify material parameters.According to the measurement error range of thermocouples,the influence of experimental measurement noise on identified results is studied.The heat transfer experiment of composite sandwich structure is designed and carried out.The integrated heat transfer model of thermal insulation material and composite sandwich structure is established.Research results indicate that the material parameter identification of the integrated heat transfer model can make it more consistent with the measurement results under the experimental conditions.The average error between the measured and simulated results of the inner temperature of the modified heat transfer model decreases from 46.0 ℃ to 1.34 ℃.Then,a heat flux identification method of composite sandwich structure is proposed based on the CGM.Based on the two-dimensional transient heat transfer model,the spatially nonuniform heat flux is identified.According to the measurement error range of thermocouples,the influence of experimental measurement noise on identification results is studied.The proposed identification method is combined with finite element analysis to identify the heat flux of the composite sandwich structure model.The heat flux for composite sandwich structure in the experimental condition is identified by using the experimental data and the modified integrated heat transfer model.Research results indicate that through the simulation analysis,the error between the identification heat flux and the reference value is small.In the experimental condition,a reasonable time curve of heat flux can be obtained.Finally,based on the Back Propagation(BP)neural network,the heat flux identification method for composite sandwich structure is proposed.The training database is established through heat transfer simulation analysis.The known temperature and heat flux information at the current time is considered as the input parameter,while the identification value of heat flux is taken as the output parameter.The BP neural network model is trained and constructed.The experimental data and the modified integrated heat transfer model are used to identify the heat flux.Research results indicate that the error between the identified heat flux and the reference value is small,and the network model has good generalization ability.From the inside temperature measurement of the structure,the accuracy of the BP neural network is slightly lower than that of the CGM but achieve much better identification efficiency.