Multi-objective optimization of thermoelectric cooler using genetic algorithms

Abstract

The thermoelectric cooler (TEC) is a kind of cooling equipment which used to dissipate heat from the devices by Peltier effect. The cooling capacity (Qc) and coefficient of performance (COP) are both significant performance parameters of a thermoelectric cooler. In this article, three-dimensional numerical simulations are carried out by finite element analysis based on the temperature-dependent materials properties. The experimental and geometrical parameters have important effects on the TEC performance which have been analysed, such as electrical current, geometric configuration of thermoelectric leg, Thomson effect, thermal contact resistances and electrical contact resistances. The results show when the Thomson effect is ignored, the maximum difference in the cooling capacity is 7.638 W while the maximum difference in the COP is 0.09. When contact effect is not considered, the maximum difference in the cooling capacity is 22.06 W while the maximum difference in the COP is 0.75. Furthermore, the cooling capacity and COP have also been simultaneously optimized according to the multi-objective genetic algorithm. The best optimal value is obtained making use of TOPSIS (technique for order preference by similarity to an ideal solution) method from Pareto frontier. Investigated on these optimal design parameters which were anticipated to provide real guidance in industry.