Prediction of energy performance of residential buildings using regularized neural models

Abstract

Human habitats are one of the major consumers of energy. Therefore, in the current age of increasing carbon footprints, analyzing energy efficiency of a building is imminent, which has been taken up in the current work. Machine learning based Artificial Neural Network-ANN approach is used in the current work to study building-energy-performance. Total eight parameters; relative compactness, surface area, wall area and roof area of the building, overall height, and orientation of the building, glazing area and its distribution are selected as the input parameters and heating and cooling loads as the output parameters. The network prediction capability was checked by comparing the predictions of the ANN architecture with the benchmark test case. A well trained and validated ANN is used to predict 96 conditions by varying glazing area and glazing area distribution. ANN is found to capture the physics efficiently. This study revealed that there is a significant potential to improve the energy efficiency of the building and the maximum saving in the cooling load can be as high as 20.67% for a fraction of the glazing areas equal to 0.15 if glazing area distribution is kept 32.5% in North, and 22.5% each in the East, South and West.