Thermoelectric (TE) devices can provide clean energy conversion and are environmentally friendly;
however, little research has been published on the optimal design of air-cooling systems for thermoelectric
generators (TEGs). The present study investigates the performance of a TEG combined with an
air-cooling system designed using two-stage optimization. An analytical method is used to model the
heat transfer of the heat sink and a numerical method with a finite element scheme is employed to
predict the performance of the TEG. In the first-stage optimization, the optimal fin spacing for a given
heat sink geometry is obtained in accordance with the analytical method. In the second-stage optimization,
called compromise programming, decreasing the length of the heat sink by increasing its frontal
area (WHSHf) is the recommended design approach. Using the obtained compromise point, though the
heat sink efficiency is reduced by 20.93% compared to that without the optimal design, the TEG output
power density is increased by 88.70%. It is thus recommended for the design of the heat sink. Moreover,
the TEG power density can be further improved by scaling-down the TEG when the heat sink length is
below 14.5 mm.