A Novel Method for Thermoelectric Generator Based on Neural Network

Abstract

The growing need for renewable energy and zero carbon dioxide emissions has fueled the development of thermoelectric generators with improved power generating capability. Along with the endeavor to develop thermoelectric materials with greater figures of merit, the geometrical and structural optimization of thermoelectric generators is equally critical for maximum power output and efficiency. Green energy strategies that are constantly updated are a viable option for addressing the global energy issue while also protecting the environment. There have been significant focuses on the development of thermoelectric modules for a range of solar, automotive, military, and aerospace applications in recent years due to various advantages including as low vibration, great reliability and durability, and the absence of moving components. In order to enhance the system performance of the thermoelectric generator, an artificial neural network (ANN) based algorithm is proposed. Furthermore, to achieve high efficiency and system stability, a buck converter is designed and deployed. Simulation and experimental findings demonstrate that the suggested method is viable and available, and that it is almost similar to the real value in the steady state with the least power losses, making it ideal for vehicle exhaust thermoelectric generator applications. Furthermore, the proposed hybrid algorithm has a high reference value for the development of a dependable and efficient car exhaust thermoelectric generating system.