Thermal sensation and comfort models for non-uniform and transient environments, Part V: Enhancements of whole-body sensation model

• Published in: Building and environment Vol. 285; p. 113562
• Main Authors: Lin, Junwei, Liang, Yan, Yang, Junran, Xie, Yongxin, Zhang, Hui, Niu, Jianlei
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2025
• Subjects: Local thermal sensation; Sensation jump; Thermal comfort; Whole-body thermal sensation; Thermal comfort; Sensation jump; Local thermal sensation; Whole-body thermal sensation
• DOI: 10.1016/j.buildenv.2025.113562
• ISSN: 0360-1323

•Logical refinement enhanced whole-body sensation model accuracy and continuity.•Inverted jumps in prediction during gradual local sensation changes were resolved.•Overcooling bias from local cooling in heat stress conditions was mitigated.•The modified model aligned prediction more closely with experimental data. The demand for predicting human responses in non-uniform and transient environments reflecting real-world indoor and outdoor conditions is increasing. To address these needs, the Center for the Built Environment at the University of California, Berkeley has developed a series of thermal sensation and comfort models. These models predict local and whole-body (overall) thermal sensations and comfort based on physiological data using logistic regression and pieced modelling techniques. However, discontinuities occur during pieced model transitions. Additionally, when the whole-body feels quite warm, the dominant influence on the whole-body thermal sensations exerted by local cooling of back, chest, and pelvis was over-estimated. This study enhances the overall thermal sensation models through systematic experimental validation and algorithmic refinement. A total of 792 valid thermal sensation surveys were collected from 42 human subjects in seven cases featuring spatial-temporal variations from indoors and outdoors. By analysing their transition patterns, pieced models are refined to improve accuracy, smoothness, and interpretability, including adjustments of a few pieced calculation models, corrections to two original smoothing functions, and mitigation of the overcooling bias in the trunk-dominated local cooling condition. The revised models achieved smoother overall sensation predictions from local thermal sensation, aligning closely with surveyed data. Combined with a multi-nodal human body, these models improve the predictive accuracy and applicability of the overall thermal sensation in dynamic and non-uniform environments and providing a foundation for further refinements in physiology-based thermal sensation and comfort models.