LAUSR

Abstract Meta-model-based optimization eases the burden of large time and computational costs by replacing the entire simulation process (which requires solutions of transient transfer equations) with the evaluation of simple meta-model functions (such as polynomials) while preserving the accuracy of dynamic simulations. The objective of this study is to present the development of meta-models to predict the cooling and heating load change of a building with variations in building envelope design factors under arbitrary climate conditions. The meta-models were developed using the results of building energy simulations for eight representative cities of climate conditions in the U.S. and were validated by comparing the predictions with simulation results for other cities in the US and Europe. The comparison showed that good accuracy was achieved with the developed unified single model.