The perception of thermal pleasure in urban outdoor spaces is crucial for enhancing winter livability. However, existing research has primarily focused on physiological thermal comfort, leaving the mechanisms by which environmental factors influence thermal pleasure unclear. There remains a lack of systematic analysis regarding the roles of psychological perception and environmental perception. This study proposes a multidimensional synergistic framework, “Street Block Built Environment-Microclimate-Psychological Perception”, to clarify the nonlinear driving mechanisms of thermal pleasure in winter urban street blocks, integrating ML with the Shapley Additive exPlanations (SHAP) method. Utilizing 1140 questionnaire datasets and concurrently measured microclimate parameters (Ta, RH, Va, SR) from six street blocks in Hefei, nine machine learning (ML) models were trained and compared. The Extra Trees (ET) model was selected as optimal (AUC = 0.968, average F1 = 87%) for interpretation. Key findings include: The contribution of psychological perception (e.g., preference degree, overall satisfaction) to thermal pleasure significantly exceeds that of microclimate parameters and built environment perception. Threshold intervals of winter microclimate were identified: SR (164.29–568.61 W/m2), Ta (2–7.70°C), Va (<0.69 m/s), and RH (32.66–58.08%) positively enhance thermal pleasure, whereas extreme values trigger discomfort. The pleasantness of the built environment is governed by the “pleasure threshold effect”. Specifically, built environment elements—including service facilities and green coverage—can only drive further improvements in pleasantness after the psychological point is reached, thereby intensifying the thermal pleasure experience. This study innovatively constructs a human-centered thermal comfort theory, providing quantitative decision support for strengthening winter street blocks’ psychological perception, optimizing microclimates, and designing spatial forms.
扫码关注我们
求助内容:
应助结果提醒方式:
