Typical years are developed from the analysis of multi-year series, selecting actual months to assembly in a single reference year, representative of the long-term typical weather. Some statistical techniques are generally involved in the development process to ensure true frequencies, sequences and cross-correlations of the weather quantities: as regard the reference year built according to the European technical standard EN ISO 15927-4:2005, TRYEN, the method is based on the Finkelstein-Schafer statistics. In this work, we exploit the same statistic with a different target: to develop an extreme reference year, ERY, by identifying those candidate months far from being representative of the long-term weather data distribution. These new artificial extreme years are composed of statistically “non-representative” months warmer in the summer and colder in the winter - which means with daily dry bulb temperature and global solar irradiation higher in summer or lower in winter than the long-term averages respectively. The analysis is performed for five Italian localities belonging to the Alpine Regions and to Sicily. Aiming to assess the efficacy of the proposed procedure, TRYEN and ERY are compared and both used to simulate the energy performance of 48 simplified buildings, parametrically built by varying insulation level, windows’ size, orientation and SHGC and kind of opaque elements.
Harvesting big data from residential building energy performance certificates: retrofitting and climate change mitigation insights at a regional scale
