Geometric optimization of thermoelectric generator using genetic algorithm considering contact resistance and Thomson effect

Abstract

Contact resistance and Thomson heat are the two major factors in the analysis of thermoelectric modules that are often being ignored. Each of these factors has an adverse effect on the output performance of a thermoelectric module. In this study, expression for maximum power output that includes both the contact resistance and the Thomson effect has been optimized using genetic algorithm to obtain the optimum geometric parameters of a thermoelectric generator. Each leg has electrical and thermal contact resistances of 2 × 10−9 Ω m2 and 1.8 × 10−4 m2 K/W, respectively. The results of the optimization for the maximum power output and the energy conversion efficiency for Skutterudites thermoelectric materials operating at a maximum temperature difference of 500 K are 30.1 W and 9.87%, respectively. When only the contact resistances are not included, the results rise by 19.4% for the maximum power output and 11.65% for the energy conversion efficiency. When only the Thomson heat is not included, the result rise by 2.66% for the maximum power output and 5.67% for the energy conversion efficiency. These two factors should always be considered in the analysis of thermoelectric modules, neglecting them can lead to an overestimation of the output performance.