Geometric optimization of segmented thermoelectric generators for waste heat recovery systems using genetic algorithm

摘要

Recently, kinds of thermoelectric (TE) materials have been applied in the waste heat recovery system. This paper further presents a geometric optimization for segmented thermoelectric generator (STEG) modules to improve the energy harvesting performance. Sixteen length ratios of two TE materials are optimized to achieve the maximum output power. In the optimization procedure, the genetic algorithm is employed to find the optimal solution, while thermal, flow, and electric fields are fully coupled and solved by the finite element method. The optimization results indicate that the maximum output power of optimal STEGs is 24.39% (parallel-flow) and 24.33% (counter-flow) higher than that of non-segmented TEGs. Besides, the STEG under parallel-flow has better electrical performances due to the increase of temperature difference and the variation of temperature distribution. The interface temperatures are just fluctuating around the turning point of ZT values, despite of the significant variations of end temperatures. The performances of each TE legs are also investigated, which shows that some TE materials have little contribution to the whole TE module. Subsequently, hybrid TEG modules are proposed to reduce the manufacturing difficulty, which are further ranked by a multiple criteria decision-making technique. Compared with optimal STEGs, two best compromise solutions decrease the maximum output power only about 1.5%, but reduce the proportion of segmented TE legs by 56.25% and 50%, respectively. (c) 2021 Elsevier Ltd. All rights reserved.