Abstract
The simulation codes validation is a crucial issue in order to improve the reliability of simulation results needed for evaluation and design purposes and for tuning the simplified quasi-steady state models of international standards. This question has been addressed defining an integrated approach based on analytical, empirical and comparative procedures such as the one included in the international framework BESTEST. This paper is aimed to extend the comparative evaluation of two already validated simulation codes, TRNSYS and EnergyPlus, by comparing the heating and cooling monthly and annual energy needs on a enlarged range of configurations obtained by combining different values of five variables (shape, orientation, insulation,
windows area and windows distribution) and three climatic contexts (Milan, Rome and Messina). The 441 annual series obtained allowed to map the relative behaviour of the codes and to characterize with inferential statistical approaches the contribution of each parameter in causing the differences.