Citation

He X, Zhang X, Zhang R, Liu J, Huang X, Pei J, Cai J, Guo F, Wang Y. Build. Environ. 2023; 228: e109866.

Copyright

(Copyright © 2023, Elsevier Publishing)

DOI

10.1016/j.buildenv.2022.109866

PMID

unavailable

Abstract

Online monitoring of thermal comfort in non-uniform and asymmetry thermal environments is essential for achieving more intelligent and efficient vehicle cabin thermal environment management. This study introduces a new approach that provides real-time feedback to a Heating, Ventilation, and Air Conditioning (HVAC) system controller concerning the occupants' thermal sensation through non-intrusive measurements. Specifically, the cabin thermal environment is divided into multiple thermal zones and a hybrid model is proposed. The hybrid model contains two sub-models, a data-driven model and a physics-based model. The data-driven model is a temperature soft sensor that non-intrusively estimates the thermal environment parameters for each thermal zone and passes the results to the physics-based model. The physics-based model can predict the overall thermal sensation of each occupant and consider the weighted effect of local thermal sensation on the overall thermal sensation. Two vehicles' cabin thermal environment data and twenty young subjects' thermal comfort votes were collected, including both cooling and heating conditions. We conclude that the hybrid model can infer cabin thermal environment parameters and thermal comfort based on limited sensors, specifically providing thermal comfort monitoring for occupants in each seating position, which has not been much investigated in previous research. In this regard, our research will contribute to broadening the consideration of individualized thermal comfort in existing vehicles.

KW: Hyperthermia in automobiles

Language: en

Keywords

Data-driven model; Occupant-centric; Physics-based model; Soft sensors; Temperature prediction; Vehicle thermal comfort