Optimization of a high-temperature thermoelectric generator using semiconductor devices via a genetic algorithm-driven design

Thermoelectric generators (TEGs) offer a promising solution for the direct conversion of thermal energy into electricity, presenting key advantages such as silent operation, compactness, and high reliability due to the absence of moving parts. However, their widespread adoption is hindered by inherently low conversion efficiencies, typically around 5% for low-temperature systems. To address this limitation, this study proposes the design and optimization of a high-temperature TEG using silicon-based semiconductor modules, with the goal of enhancing both efficiency and output power for distributed generation applications. The proposed system incorporates a quasi-static Maximum Power Point Tracking (MPPT) algorithm for initial configuration, followed by multi-parameter optimization using a genetic algorithm (GA) to maximize performance. A comprehensive multi-physics modeling framework is developed to evaluate the thermal, electrical, and geometric characteristics of the module under idealized conditions. However, these results are based on assumptions such as perfect thermal contact and uniform material properties, which may differ in practical, real-world conditions. Comparative analyses demonstrate that the GA-optimized design results in substantial improvements in both energy conversion efficiency and output power compared to the baseline configuration. These findings underscore the predicted potential of optimized TEGs for integration into high-gradient thermal environments, such as industrial waste heat recovery and distributed power systems, based on numerical simulations. The results presented here are based on numerical multi-physics modeling and GA optimization. While the simulations show promising results, experimental validation is necessary to confirm these predictions in real-world conditions. Future work will focus on this experimental validation.