Uncertainty and Sensitivity Analysis for Building Energy Rating

Abstract

One of the main tasks of the European Energy Performance of Buildings Directive 2002/91/EC (EPBD) pertains to energy building certification. For this purpose, recent European standards have defined a methodology for the assessment of the energy use of buildings through the calculation of energy ratings. The accuracy of the energy rating depends both on the detail level of the model and on the confidence of the input data. In this work, the quasi-steady simplified monthly method, presented in the ISO 13790: 2008 standard, has been applied to a dwelling in Turin, Italy, in order to calculate its energy rating. A widespread list of 129 input data was identified and grouped into three sets: climatic data, envelope data, and building use data. Probability density functions were assigned after a detailed bibliographic research. The Monte Carlo Latin hypercube sampling technique has been used to assess the building energy needs confidence interval and the energy performance class uncertainty. A sensitivity analysis based on the Morris method, has also been performed for the different building heat balance terms, in order to identify the most important parameter set that takes into account for the uncertainty in the model output. The main results demonstrate that the asset energy rating is only slightly affected by data uncertainties; moreover, the analysis highlights the importance of associating an operational energy rating to the asset rating in order to show the influence of user behavior on building energy performance. On the other hand, the results show that less than 10 input data, among the considered 129 ones, have a significant influence on the energy rating uncertainties.