Accurate measurement of thermal conductivity parameters of materials is of great significance in modern industrial field.In recent years,transient plane heat source method has been widely used because of its characteristics of fast measurement speed,high measurement accuracy and wide measurement range.But the TPS standard method is measured in a laboratory.It is necessary to prepare the sample to meet the experimental requirements in advance.It is not possible to measure wall,ground and other materials that are not easy to obtain standard test samples.Therefore,it is of great practical significance and application value to develop a portable method for measuring thermal conductivity parameters.In order to solve the problem that the existing thermal conductivity measurement method can not achieve real-time field measurement,the TPS standard method has improved.The measurement condition that requires two samples to be tested is changed to one single sample to be tested.The improved method is portable and easy to operate,and can be applied to a variety of measurement scenarios.It provides an effective measurement method for obtaining standardized sample materials.Based on the theoretical basis of TPS standard method,the transient heat transfer problem under the semi-infinite plate model was studied,and the formula for measuring thermal conductivity of single-side TPS method was derived,and the corresponding probe heat capacity correction model was established.In order to illustrate the theoretical feasibility of single-sided TPS,the influence factors of single-sided TPS method were explored,the finite element simulation of the measurement process of single-sided TPS method was carried out,the influence of different heating power and different thermal diffusion coefficient on the measurement results of different materials was analyzed,and the time delay problem existed in the actual measurement was discussed.The results show that single-side TPS method has good measurement effect theoretically and is feasible.In order to explore the practical measurement effect of single-side TPS method,this paper developed a measuring equipment of single-side TPS method,respectively measured the transient temperature changes of three standard materials,including 45 steel,SUS304 stainless steel and 40 Cr,under different heating power,and used single-side TPS method to calculate the thermal conductivity of the data.The results show that the measurement error of single-sided TPS method is less than 10%,which further shows that single-sided TPS has good practical significance.In order to solve the problem that the probe thermal capacity and void thermal resistance of the TPS method are difficult to eliminate,this paper explores the feasibility of combining the single-side TPS method with the neural network model.Based on the finite difference model,the mathematical heat transfer model of single-side TPS method is established,and the transient temperature rise database of materials with different thermal conductivity parameters is established through this model,which provides training data for the neural network model.The actual measurement data is used as the input of neural network prediction to predict the thermal conductivity parameters of the measured materials.The influence of different data lengths,time delays and neural network models on the prediction results is explored,and the sensitivity of the neural network prediction scheme is analyzed.The results show that the combination of single-sided TPS method and machine learning model is feasible,and the accuracy of the prediction effect depends on the perfection of the database and the standard of measurement. |