Research on the optimization of building energy efficient measures based on incremental costs

Abstract

Based on the incremental costs analysis, this paper optimizes the energy efficient measures for buildings with the prediction of building energy consumption benefitting from 18 building envelope performance parameters by using artificial neural networks. A BP neural network has been preferred and the data have been presented to network by being normalized. The building energy simulation software DeST was used for the calculations of energy consumption and ANN toolbox of MATLAB is used for predictions. Then five combinations of these materials for the building were obtained by the predictions of building cooling and heating energy consumption with this BP neural network for the purpose of getting the same energy efficient rate. Results show that BP neural network gives satisfactory results with successful prediction rate of over 98% at early design stage and provides a fast method to optimize building energy efficient measures to reduce incremental cost.