Physiological sensing for thermal comfort assessment

Accounting for inter- and intra-personal differences requires individual and cohort comfort models. For their development, emulators for thermal sensation of occupants are needed. Physiological signals can be acquired using both wearable and contactless devices. However, due to the widespread availability of sensing methods it is difficult to select the proper measuring method for the application. The objective of this study is to provide an overview of the capabilities of contemporary devices that measure physiological indicators used in literature and identify their capabilities and limitations. The analysis was made on a dataset of reviewed thermal comfort research studies that employed physiological sensing devices in experimental and field test campaigns. The physiological indicators investigated in literature were derived from the human thermoregulation mechanism. The physiological indicators measured were neural activity (brainwave frequency bands), heartbeat (heart rate and heart rate variability), blood flow (blood pressure, blood oxygen saturation, skin blood flow), activity (metabolic rate, activity, calorie consumption), temperature (core and skin), sweat (relative humidity, skin conductance, skin hardness, and amount of sweat). The wrist is the most investigated body part as it is a convenient area for acquiring multiple physiological indicators i.e., all physiological measurements except for ECG and EEG measurements. However, most devices are not “plug-and-play” solutions for thermal comfort assessment. As contact devices, smartbands acquire an extensive set of indicators but present 3rd party data privacy protocols which may limit their applicability. Cameras (RGB and infrared) can only be used to acquire skin temperature and heart rate but can be deployed in the space by the building owner. Further studies are required on the sensing accuracy and signal variability as a function of thermal sensation to determine the optimal measurement method.