Predicting the effective thermal conductivity of dry granular media using artificial neural networks

Marcin Grabarczyk, Piotr Furmański


Thermal conductivity of heterogeneous materials is a complex function not only of properties and amounts of constituents but also of many structural factors. Therefore it is difficult to predict. An attempt to predict thermal conductivity of granular media using  the Artificial Neural Network (ANN) model is undertaken in the paper. It was assumed that it is a function of a ratio of thermal conductivities of the constituents, medium porosity as well as the coordination number describing the mean number of the nearest neighbours to each grain. Several configurations of the ANNs were tested while developing the optimal model. As a measure of prediction accuracy the coefficient of linear regression and the mean squared error were used. The optimal model of ANN was found to consist of three hidden layers with eight neurons in each layer for both types of media. Some problems associated with application of ANN were pointed out. The predicted values of thermal conductivity obtained with ANN were compared with values calculated from an analytical formula. It was found that the ANN predictions show identical trends and similar values as the analytical formula for all factors affecting thermal conductivities of the granular media.


granular materials; thermal conductivity; prediction; artificial neural network

Full Text:



Bishop M.C.: Neutral network and their applications. Review in Scientific Instruments 64, 1994, s. 1803–1831.

Chayjan R.A., G. A. Montazer, T. T. Hashjin, M. H. Khoshtaghaza, B. Ghobadian: Prediction of pistachio thermal conductivity using Artificial Neural Network approach. International Journal of Agriculture & Biology 9, 2007, s. 816-820.

Crane R.A., Vachon R.I.: A prediction of the bounds on the effective thermal conductivity of granular materials. International Journal of Heat & Mass Transfer 20, 1977, s. 711-723.

Dayhoff J.E.: Neural Networks Principles,Prentice-Hall International, 1990.

Fayala F., Alibi H., Benltoufa S., Jemni A.: Neural Network for predicting thermal conductivity of knit materials. Journal of Engineered Fibers and Fabrics 3 (4), 2008, s. 53-60.

Furmański P., Wisniewski T.S., Łapka P.: Study on a degree of degradation of design materials under different ambient interactions. In Kurnik W. (editor): “Nonconventional materials in diagnostics and active reduction of vibration”, Scientific Publications of Institute of Exploitation Technology, Warsaw - Radom 2008, s. 207-233 (in Polish).

Gemant A.: The thermal conductivity of soils. Journal of Applied Physics 21, 1950, s. 750-752.

Gogół W., Próchniak A.: Heat transfer in granular media. Bulletin of Institute of Heat Engineering, Warsaw University of Technology,1978, s.

Gori F., Corasaniti S.: Theoretical prediction of the soil thermal conductivity at moderately high temperatures. ASME Journal of Heat Transfer 124, 2002, s. 1001-1008.

Goutorbe B., Lucazeau F., Bonneville A.: Using neural networks to predict thermal conductivity from geophysical well logs. Geophysical Journal International 166 (1), 2006, s. 115–125.

Hornik K., Stinchcombe M., White H.: Multilayer feedforward networks are universal approximators. Neural Networks Comput. 2(5), 1989, s. 359–366.

Hussain A.M., Rahman M.S.: Thermal conductivity prediction of fruits and vegetables using neutral networks. International Journal of Food Properties 2, 1999, s. 121–138.

Khanna T.: Foundations of Neural Networks. Addison-Wesley Publishing Company, 1990.

Sablani S.S., O-D. Baik, M. Marcotte, Neural networks for predicting thermal conductivity of bakery products, Journal of Food Science 52, 2002, s. 299-304.

Stefański A.: Thermal conductivity of civil engineering materials. Polish Scientific Publisher PWN, 1975(in Polish).


  • There are currently no refbacks.