Machine learning-based personal thermal comfort model for electric vehicles with local infrared radiant warmers

Thermal comfort of occupants in conventional vehicles driven by an internal combustion engine is controlled by heating, ventilation, and air conditioning (HVAC) system. However, the operation of a conventional HVAC system decreases the mileage of electric vehicle (EV) in the heating mode by nearly 5...

Full description

Saved in:
Bibliographic Details
Published inJournal of mechanical science and technology Vol. 35; no. 7; pp. 3239 - 3247
Main Authors Lee, Yein, Lee, Hyunjin, Kang, Byung Ha, Kim, Jung Kyung
Format Journal Article
LanguageEnglish
Published Seoul Korean Society of Mechanical Engineers 01.07.2021
Springer Nature B.V
대한기계학회
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Thermal comfort of occupants in conventional vehicles driven by an internal combustion engine is controlled by heating, ventilation, and air conditioning (HVAC) system. However, the operation of a conventional HVAC system decreases the mileage of electric vehicle (EV) in the heating mode by nearly 50 %. Thus, local radiant heating was proposed as a heating strategy to reduce electric energy consumption while providing reasonable thermal comfort. In this work, we developed a personalized overall thermal sensation (OS) model using machine learning to evaluate the thermopsychological effect of local radiant heating and simulate the OS of occupants in EVs. Data were obtained from a real EV that went through a cold environmental chamber and were evaluated using random forest algorithm. By considering individual thermal preferences of occupants, we predicted the OS for each subject with higher accuracy by a factor of 2.6 compared with the prediction performed using the weighted average method. Total energy consumption was reduced by approximately 10 % in the EV equipped with local infrared radiant warmers while providing OS that was comparable with that of the HVAC system.
ISSN:1738-494X
1976-3824
DOI:10.1007/s12206-021-0644-7