Assessing and enhancing the applicability of standard effective temperature in outdoor thermal sensation prediction for continental climates

Abstract

Accurate thermal sensation prediction is crucial for assessing outdoor thermal environments. This study aims to enhance the applicability of the standard effective temperature (SET*) in continental climates for a more comprehensive evaluation of outdoor thermal comfort. Taking Zhengzhou, China as an exemplar, a comprehensive 12-day survey involving questionnaire surveys and microclimatic measurements was conducted in residential communities during the summer, autumn, and winter seasons. Based on this, the relationship between the SET* and thermal sensation was explored. Furthermore, the SET*-original thermal sensation scale was modified using linear and probit regression, and the accuracy of modified scale was assessed using predicted accuracy percentage (PAP) and mean absolute error (MAE). The results indicated that the SET*-original scale performs poorly in predicting thermal sensation in continental climates (P = 27.3 %, MAE = 1.24). The modified scale using probit regression model was better suited for continental climates (P = 39.5 %, MAE = 0.87). It outperformed the modified scale obtained through a linear regression model, with PAP increasing by 5.9 % and MAE decreasing by 0.04. The SET*-modified scale extended thresholds at the colder end and significantly enhanced predictive accuracy for colder conditions, making it more suitable for predicting thermal perception in continental climates.