Performance optimization of a thermoelectric generator for automotive application using an improved whale optimization algorithm

Due to the low conversion efficiency and limited figure of merit (ZT) of thermoelectric modules (TEMs), the output power of thermoelectric generators (TEGs) should be improved for automotive exhaust waste recovery in addition to considering its impact on engine. In this study, a polyhedral heat exchanger was utilized to construct a TEG system and harvest the exhaust heat; its thermal behavior and power generation performance were assessed with a multiphysical field coupling model, and a Gaussian process regression (GPR) orthogonal model was established. Finally, the backpressure, conversion efficiency, temperature distribution, and temperature uniformity of TEG were considered to design the optimization objective, length, width, angle, and spacing of fins inserted inside the polyhedral heat exchanger, which were selected as design variables, and the overall performance of the TEG system was optimized with an improved whale optimization algorithm (IWOA). The results demonstrate that in contrast to the TEG system using an initial polyhedral heat exchanger without optimization, the mean temperature, backpressure, conversion efficiency, and peak power optimized with IWOA is decreased by 0.74%, 15.27%, 3.58%, and 9.4%, respectively, and the temperature uniformity coefficient and optimization objective is increased by 1.30% and 11.67%, respectively. Moreover, the one optimized with IWOA has obvious advantage over the one optimized with ordinary optimization whale algorithm (WOA), as its optimization objective is increased from 0.912 to 0.947, which is increased by 3.84%. The proposed method is feasible as it well balances the power generation and backpressure, and the optimized TEG system is more suitable for automotive application.