Integrated personalized thermal comfort model for input variable reduction

Chuangkang Yang, Keiichiro Taniguchi, Shohei Miyata, Yasunori Akashi

Proceedings of the 10th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation 271-273, November 2024

https://doi.org/10.1145/3600100.3626256

Abstract

To improve indoor overall thermal comfort and reduce building energy consumption, one of the methods is building a personalized thermal comfort model which can predict individual thermal sensation and thermal preference. Although a number of models have been proposed previously, they have significant drawbacks that limit their use in daily life. This study proposes a model which combines mathematical modeling and machine-learning algorithms. It offers personalized thermal sensation values for each individual based on measured wrist skin temperature, heart rate, and indoor body ambient air temperature. Compared with previous studies, the proposed method uses fewer input variables and improves prediction accuracy.