Three-zone numerical modeling method for predicting system-level waste heat recovery performance of thermoelectric generator with various electrical array configurations

In this study, an experimentally validated numerical model is formulated to accurately predict the energy harvesting performance of a thermoelectric generator (TEG) by reflecting the underlying physics of the thermoelectric energy conversion phenomena. The most advantageous feature of the modeling method employed herein is its high reliability because the system-level performance is predicted based on module-level power generation characteristics. A further benefit that the proposed physics-based model affords is that the effect of electrical array configurations (i.e., the particular manner by which the thermoelectric modules (TEMs) are electrically connected) on the waste heat recovery performance of the TEG can be accurately predicted. For this purpose, a three-zone modeling method, wherein each TEM is modeled into heat sink, heat generation, and heat conduction zones, is proposed. Moreover, user-defined functions are embedded into the numerical model to correlate the temperature fields obtained from the transport of momentum and energy with the Seebeck effect and Joule heating that appear as the thermoelectric output power and heat sink/heat generation near the TEM surface. The accuracy and reliability of the numerical model are validated by the experimental results of temperature and output power using a TEG prototype whose configurations and dimensions are identical to those of the proposed numerical model. Further comparisons between the numerical results obtained using eight different electrical array configurations and the corresponding experimental results ensure the robustness of the modeling method and numerical model formulated in this study. Finally, two correlations are derived as a function of the number of electrical array branches to predict the output voltage and current of a TEG whose branches consist of an identical number of TEMs for minimizing unexpected power losses.