Data for: Framework for the ISO 52016 standard accuracy prediction based on the in-depth sensitivity analysis
Description
The data accompanies the article "Framework for the ISO 52016 standard accuracy prediction based on the in-depth sensitivity analysis" which presents the accuracy of ISO 52016-1:2017 standard (the Standard) compared to numerical dynamic simulations in energy modeling software TRNSYS. The comparison has been done for hourly and annual heating and cooling energy needs for all international climate zones defined in the ASHRAE Standard 90.1, 2 building geometries and a huge variety of building properties. The differences between the Standard and TRNSYS have been quantified using the root mean square error (RMSE) and the coefficient of variation of root mean square error (CVRMSE) based on the hourly energy needs, as well as the relative error (RE) based on the annual energy needs. The energy needs have been calculated for a wide range of building properties, resulting in 4608 analyzed cased. Input values for each of those cases can be found in the folder “Input parameters”. The annual heating and cooling energy needs calculated by TRNSYS can be found in the folder “Annual energy needs”. Calculated differences between TRNSYS simulations and the Standard can be found in the folders “CVRMSE”, “RMSE“ and “RE”. A global sensitivity analysis (using Sobol method) has been performed, resulting in sensitivity indices, which can be found in the folder "Sobol indices". Each folder contains a description file regarding the data. The document “Figures.docx” gives a full set of relevant figures, some of which have not been presented in the original paper for brevity reasons. Any additional inquiry about the data set can be sent to the corresponding author, prof. Tea Zakula (tzakula@fsb.hr).