Transient Volumetric Heat Source Method Based On Direct Numerical Simulation Of Physical Properties Of Thermal Conduction

Thermal conductivity and thermal diffusivity,as important parameters of thermal properties,are important indexes reflecting the heat transfer performance of materials,and provide important basis for the design of heat transfer involved in all walks of life.With the continuous discovery of new materials and the continuous expansion of the conditions for use of materials,it is of great significance for the accurate measurement of thermal conductivity and thermal diffusivity of materials.After years of research,a variety of methods have been developed to measure thermal conductivity and thermal diffusivity.Different methods can be used depending on the measurement object,measurement range,and measurement accuracy.Among them,the transient heat source method has the advantages of small instrumentation,short measurement time,accurate measurement,etc.,and has become the main research direction.The transient heat source method is typically represented by the hot line method and the transient planar heat source method.Since it is difficult to obtain the analytical solution of the temperature response of the probe when considering the volume and exact shape of the heat source,the volume of the heat source is not considered during the establishment of the ideal model of the two methods,and the heating wire is considered as a wire without thickness and without a diameter.Considered infinitely thin,and even with corrections,the volμme of the heat source cannot be accurately considered.Therefore,the probe heat capacity,probe size,change in heating power,contact thermal resistance,time delay and other factors have become sources of error.In order to take the volume of the heat source into account in the theory of the measurement method,the errors caused by factors such as the head heat capacity,the probe size,and the contact thermal resistance are reduced.This paper proposes a method for inversion of thermal conductivity and thermal diffusivity based on numerical calculations.Because the mathematical expression of the temperature response of the probe cannot be accurately obtained when considering the volume of the heat source,a database needs to be established to perform data fitting.Calculate the temperature response of the probe under different conditions in the positive problem and establish a non-dimensionalized database.In the inverse problem,based on the temperature response of the probe,the database is used to invert the thermal conductivity and thermal diffusivity of the tested material.In this method,since the accurate shape and volume of the probe are taken into account when simulating the heat transfer process in the positive problem,the probe size,heat capacity,and sample boundary will not introduce systematic errors,so the accuracy can be improved;and because the inside of the probe Embedded in the detector base with known thermal properties not only can reduce the contact thermal resistance,but also does not require the probe to meet the ultra-thin conditions,which is more conducive to the construction of physical objects.In this paper,the program simulates the heat conduction process and verifies the correctness of the program through examples.A database was created using the program and the process of correcting the problem was completed.The neural network was used to train the data and the neural network model was obtained.The temperature response of the probe was predicted.The neural network model and the simplex search method are used to program the thermal properties of the tested material.In other words,the thermal conductivity and the thermal diffusivity of the tested material can be calculated accurately by using the known temperature response of the probe.According to the results of the inversion,it can be seen that the scheme can accurately invert the thermal conductivity of the measured material,and from the perspective of relative error,the degree is higher than the inversion accuracy of the thermal diffusivity.Among them,the absolute error between the inverse thermal conductivity coefficient and the true value is about 1 W/(m K);the absolute error between the inverse thermal diffusivity and the true value is within 5 20.3 10 m /s-?.