An efficient performance assessing index for building climatic zoning methods based on probability density functions

Abstract

Climatic zoning is the basis of improving thermal performance of buildings, and various methods and parameters have been proposed in previous studies. However, the validity assessment of these climatic zoning methods is rather limited. This study proposes a performance index (PI) for assessing climatic zoning methods by comparing the overlapping probability density functions (PDFs) among different climate zones. The PI was used to assess the performance of the two subjective (i.e., based on the two national standards GB50178 and G50176) and one objective (i.e., cluster analysis) methods of climatic zoning in China with over five different climate zones. The results showed that the cluster analysis method has the smallest mean PI (16.8) compared with those of the subjective methods (17.8 and 19.3). More importantly, the cluster analysis method has less variations in PIs among relevant variables (13.8–23.2), while the PIs largely ranging from 6.4 to 42.7 and 10.4 to 29.9 for the two subjective methods. These results indicate that the PI can accurately assess the validity of climatic zoning methods, and that the objective method may be better for building climatic zoning for the whole China as it can effectively distinguish the relevant climatic variables among different climates. The quantified index may have the potential to assess other zoning methods, particularly when applied in countries with larger areas and more climate zones.