Estimation of outdoor design dataset based on interdependency of multiple meteorological elements by using vine copulas

In building design, the commonly used outdoor weather conditions recommended by ASHRAE are gen-erated by the annual cumulative frequency of occurrence method. This may result in air-conditioning equipment capacity being overestimated. This is because the air temperature, humidity and solar radia-tion that have the greatest impact on indoor peak cooling loads are selected independently. It means that the probability of these three weather parameters occurring at the same time is extremely small. This paper proposes an approach for selecting the design weather data on the basis of the exact joint distri-bution with specific simultaneous occurrence probabilities of these three weather parameters. For this purpose, a novel statistical method vine copula is applied to construct the theoretical multivariate joint distributions of these three weather parameters for 24-hour data and combined the design parameters of each hour with a specific simultaneous occurrence probability into a design day. The hourly meteorolog-ical records over 10 years in Tokyo are used to generate the outdoor design conditions. The results indi-cate that the vine copula method can create an effective theoretical model and all the design parameters selected by the simultaneous occurrence probability are significantly below that of the same conven-tional exceeding probability. For simultaneous occurrence probabilities of 0.4 %, 1 % and 2 %, the corre-sponding indoor thermal risk levels are 0.46 %, 0.52 % and 0.84 %, respectively. Therefore, the proposed method can be used as a reference for designers to determine a more reasonable capacity of air-conditioning systems.(c) 2022 Elsevier B.V. All rights reserved.