Estimating the Thermal Bridge at Wall Corners with Artificial Neural Network

Abstract

Nowadays there are several software products to simulate the heat transfer at thermal bridges.  These sophisticated tools are mostly based on Finite Difference or Finite Element Methods. However, there is a demand from energy modelers to estimate the linear thermal transmittance and the lowest internal surface temperature from existing data, instead of employing time-consuming simulations for each case.  This paper investigates wall corners with and without thermal insulation by using Artificial Neural Network (ANN), as a Soft Computing technique. It gives simple formulas that estimate the thermal bridge at wall corners with an acceptable level of accuracy.