A holistic 3D finite element simulation model for thermoelectric power generator element

Harvesting the thermal energy stored in the ambient environment provides a potential sustainable energy source. Thermoelectric power generators have advantages of having no moving parts, being durable, and light-weighted. These unique features are advantageous for many applications (i.e., carry-on medical devices, embedded infrastructure sensors, aerospace, transportation, etc.). To ensure the efficient applications of thermoelectric energy harvesting system, the behaviors of such systems need to be fully understood. Finite element simulations provide important tools for such purpose. Although modeling the performance of thermoelectric modules has been conducted by many researchers, due to the complexity in solving the coupled problem, the influences of the effective Seebeck coefficient and carrier density variations on the performance of thermoelectric system are generally neglected. This results in an overestimation of the power generator performance under strong-ionization temperature region. This paper presents an advanced simulation model for thermoelectric elements that considers the effects of both factors. The mathematical basis of this model is firstly presented. Finite element simulations are then implemented on a thermoelectric power generator unit. The characteristics of the thermoelectric power generator and their relationship to its performance are discussed under different working temperature regions. The internal physics processes of the TEM harvester are analyzed from the results of computational simulations. The new model presented by this paper is more conservative than the traditional model in predicting the performance of TEM power generator working under strong-ionization region. By accounting for the mechanisms that are typically ignored or simplified in the existing models, this new model provides more holistic descriptions on the thermoelectric power generator behaviors and therefore potentially will further improve the accuracy in the computational simulations.