화학공학소재연구정보센터
Energy Conversion and Management, Vol.180, 584-597, 2019
An approximate and efficient characterization method for temperature-dependent parameters of thermoelectric modules
Thermoelectric generator which converts waste heat to electricity is a promising approach in energy field. In this paper, a comprehensive numerical model of a thermoelectric couple is built by COMSOL with all related thermoelectric effects and some irreversible effects considered. It is adopted instead of a whole thermoelectric module modeling for a less calculation demanding and time consumption purpose, whose feasibility and equivalence are verified as well. An easy-manipulative symmetric test rig is designed to precisely acquire the electrical and thermal variables of thermoelectric module in terms of heat flow rate, voltage and current in a wide temperature range. The good agreement of test and simulation results validate the effectiveness of the numerical model and the feasibility of test device including its control and measurement systems. Then a novel characterization method called quasi-steady-state method is first proposed to fast and precisely derive the module-level temperature-dependent thermoelectric parameters including Seebeck coefficient S, thermal conductivity lambda and electrical resistivity rho. The characterization procedure is to raise the temperature of the thermoelectric modules hot side at an appropriate heating rate and keep the chiller at a constant temperature both in open and short circuit tests in sequence. The thermoelectric leg is regarded as multiple serial segments divided according to its temperature distribution. One set of equations involving the thermo- and electro- variables is established for each segment in case of a certain temperature difference. All these equations are solved by means of an iterative algorithm and computed by MATLAB. S and lambda are obtained by open circuit tests and rho by the short circuit tests subsequently. The comparison of estimated thermoelectric parameters by tests, simulations and the manufacturers data validate the correctness of the quasi-steady-state method and its predictability ability of the performance of thermoelectric generators in practice.