Investigation of the underground temperature using neural network

Abstract

In this paper, we use the concept of neural networks to evaluate the underground temperature of any type and depth of earth. Two models have been developed for this purpose namely mathematical and intelligent. The mathematical model is developed taking into account properties of the soil and meteorological conditions, whereas the intelligent model is a development of data driven neural network model. Fourth variables influencing the underground temperature which were taken into account are ambient temperature, underground depth, soil thermal diffusivity and days of year. The model was validated against experimental data sets.