An electrical analogy model for performance evaluation of natural air-cooled TEG modules considering nonlinear effect of heat convection

Abstract

With the outstanding merits of the simple principle and low computational cost, the electrical analogy model of thermoelectric generator (TEG) modules plays an important role in evaluating the performance of thermoelectric energy harvesting systems. However, owing to the simplification of the heat dissipation structure and difficulty in determining the heat dissipation parameters, it is difficult to accurately describe the nonlinear influence of thermal convection on the heat dissipation performance of the TEG module under natural air-cooled conditions, thereby reducing the estimation accuracy of the module temperature and output. A new nonlinear electrical analogy model to evaluate the performance of natural air-cooled TEG modules was proposed in this study. The proposed model more accurately describes the nonlinear influence of thermal convection on the thermal dynamic characteristics of the TEG module by introducing a heat transfer branch between the cold and hot side surfaces of the TEG module and the environment. A parameter extraction method based on multi-objective optimization and the non-dominated sorting genetic algorithm II (NSGA-II) was proposed to overcome the problem of the heat dissipation parameters of natural air-cooled TEG modules. The experimental results show that under natural convection conditions, the maximum estimation errors of the temperature and output voltage evaluated by the proposed model and parameter identification method are less than 6.18 % and 9.14 %, respectively. Under forced convection, the maximum estimation errors of the temperature and output voltage were less than 7.36 % and 8.71 %, respectively. The above estimation errors are significantly lower than the traditional model.