Refining adaptive models for overheating evaluation: Evidence-driven comparative analysis of residential buildings in China's cold and severe cold regions.

China's residential buildings in cold and severe cold regions grapple with unaddressed summer overheating issues owing to the lack of tailored regional standards, necessitating a critical examination through globally recognized thermal comfort benchmarks. In this study, 27 bedrooms and 47 living rooms in 13 representative cities were monitored from May to September in the years 2021–2023 to investigate the summer indoor operative temperature of residential buildings in 5 sub-regions. The accuracy of adaptive models from ASHRAE 55, EN 16798–1, and GB/T 50785 for overheating evaluations was tested using measured data. Further, the Chinese Thermal Comfort Database (CTCD) was expanded in this study to enable the development of a preliminary Summer Residential Adaptive Model (SRAM) specifically designed by considering the distinct characteristics of China's cold and severe cold regions. Analysis of the monitoring results showed that 73.1 % of living rooms and all bedrooms were overheated according to CIBSE TM59. Moreover, the slope of SRAM is 0.40, which is lower than 0.73–0.77 of GB/T 50785 and higher than 0.31 and 0.33 of ASHRAE 55 and EN 16798–1, respectively. Further comparative assessments revealed inconsistencies with actual human sensations of heat. The results of this study highlight the inadequacy of CIBSE TM59 and existing thermal comfort standards in accurately assessing building overheating in China's cold and severe cold regions. To improve evaluation, future research must integrate regional climate modeling, various architectural configurations, and thermal sensation indicators to refine the developed SRAM, thus enabling the establishment of more holistic thermal comfort standards. • Highlights standard gaps in cold climates by documenting widespread residential overheating across 74 rooms in 13 Chinese cities. • Refines an adaptive model (slope 0.40) using 21 thermal comfort datasets from 5 sub-regions. • Exposes CIBSE TM59 and standard thermal comfort models' limitations in evaluating overheating accurately. [ABSTRACT FROM AUTHOR]