A comprehensive analytical model for thermoelectric body heat harvesting incorporating the impact of human metabolism and physical activity

Wearable electronic devices, which can provide hassle-free, long-term, continuous monitoring for different health parameters are of interest for a variety of applications including managing chronic diseases and post-operative patient care. Thermoelectric generators (TEGs) that rely on the Seebeck effect provide a promising path to self-powered wearable electronics via harvesting body heat. The net harvested energy depends on several factors including metabolic heat, core body temperature, and skin resistance, which are regulated by the human thermoregulatory system. These factors are influenced by age, sex, height and body mass index (BMI) and they can cause appreciable differences in the amount of harvested energy. In this paper, we present a comprehensive model, which combines an analytical TEG model with established biological models for the human body to accurately predict the performance of a wearable TEG. The model calculates variations in metabolic rate and core body temperature during physical activity, which are then used to calculate the skin resistance and the skin temperature following an iterative procedure. These parameters are then used to determine the temperature differential across the TEG and the resulting output power delivered to an external load. Experimental validation of the model was achieved using a wrist worn flexible TEG during different physical activities, which were carefully designed to separate the impact of convection from the human thermoregulatory response. It was determined that increase in metabolism from normal walking could lead to a 20% increase in the TEG output voltage. The model was used to predict the impact of age and sex on TEG output power. The results indicate that age can have a significant impact on TEG performance. It is shown that the power generated by adults over the age of 65 can be 30%–35% less than the power generated by younger adults under the age of 30. This reduction in power was attributed to an increase in skin resistance, which was found to be 13% for females and 25% for males. The skin resistance of females was also found to be higher, which correlated well with their higher average fat content. The overall impact of sex was found to be smaller than age, females generating 5%–10% less power than males. The model takes into account all pertinent TEG parameters including properties of the semiconductor materials, physical dimensions of the semiconductor legs, fill factor and electrical and thermal parasitic resistances. Using these input parameters, and modeling of the human body as a heat source, the model can help optimize TEG module design for a specific wearable application.