Reconfiguration of Thermoelectric Generation Systems based on Artificial Bee Colony Algorithm

Abstract

Thermoelectric generation (TEG) is regarded as one of the most promising green power generation methods. However, the complex and changeable environment often causes the temperature difference array to operate under the condition of non-uniform temperature difference (NTD), the output power is low, and the power-voltage (power-voltage, P-V) curve has obvious multi-peak characteristics, which is not conducive to the system stable and efficient operation. Traditional strategies are difficult to completely suppress and eliminate this adverse effect, so a modular temperature difference array reconstruction technology based on artificial bee colony (ABC) is proposed, which aims to fundamentally solve the problem of various uneven temperature distributions to TEG system. Firstly, a (15×15) modular TEG array reconstruction model is built by MATLAB/Simulink, the entire TEG array is divided into three array modules, and various dynamic reconstructions can be completed only by configuring two switch matrices, reducing the construction and operation costs of TEG system. Then, ABC is guided to perform reconstruction optimization with the goal of maximizing output power. The simulation results show that ABC can effectively smooth the P-V curve, increase the output power, reduce the voltage unbalance, fully tap the power generation potential of TEG array, and ensure the safe and stable operation of the system.