Thermal Sensation Modeling and Experiments for Liquid-Cooled Garments

Abstract

Liquid-cooled garment (LCG) is a promising personal thermal management (PTM) technique that satisfies the individual thermal comfort requirement by creating a microclimate around the human body. Thermal sensation, which is defined as “the wearer’s sense of temperature between hot and cold,” is a helpful target for designing an LCG with good thermal comfort. There are just a few methods to predict the thermal sensation of the LCGs. In addition, the previous prediction methods related to the conventional heating, ventilation, and air conditioning (HVAC) systems cannot be directly applied to the LCG, due to the lack of consideration of some dominating characteristics of the LCG and the human body. To solve this problem, a neural network model was proposed to predict the thermal sensation of wearers of LCGs, taking physiological parameters [heart rate (HR), skin temperatures, and tympanic temperature] and physical parameters [ambient temperature (AT) and relative humidity (RH)] including water inlet temperature of LCGs into consideration. Experiments were carried out to obtain the model training data under various conditions. After optimization, the neural network model performs excellently, which shows the potential to predict the thermal sensation for LCGs. The correlation analysis indicates that water inlet temperature is the most correlated parameter to thermal sensation in LCGs.