An artificial neural network-based prediction of government-owned building energy consumption with design variables

Abstract

An accurate prediction of the energy consumption of buildings in the design phase is vital for organizations, as it helps to formulate early phases of development to reduce the environmental impact of such buildings. Accurate model is needed to gauge the energy consumption prediction of government-owned buildings in the design phase. The aim of this study is to predict energy consumption of government-owned buildings by considering 26 variables, which are defined in the design phase using artificial neural network (ANN) method. The proposed ANN method analyzed and validated 175 sets of data derived from the 2003 CBECS database. Additionally, the result obtained using the proposed ANN model was compared with multiple linear regression (MLR) method. Experimental results revealed that the proposed ANN model is able to predict the energy consumption of government-owned buildings in the design phase. © 2013 American Society of Civil Engineers.