Overall heat loss coefficient and domestic energy gain factor for single-family buildings

In this work we introduce a method for estimating the variation of the overall heat loss coefficient and the domestic energy gain factor. Based on a neural network model, these parameters were extracted from analyzing the model by indirect methods. The used model parameters were: the supplied heating demand, the domestic electrical demand and the indoor–outdoor temperature difference. A feed-forward back propagation neural network was used as modeling tool. The proposed method has been found accurate, based on an analysis of artificially generated data. Additionally, measured data of inhabited single family buildings were examined and the model was found to generate reliable results, in parity with results obtained by comparable methods and estimations.