Improved immune genetic algorithm based TEG system reconfiguration under non-uniform temperature distribution

To deal with the problem of low thermoelectric conversion efficiency in thermoelectric power generation (TEG), this work designs an improved immune genetic algorithm (IIGA) to reconfigure TEG system under non-uniform temperature distribution (NTD) conditions to maximize power output. The traditional IGA is a fusion algorithm that combines genetic algorithm (GA) with immune algorithm (IA). Due to IGA's flaws of strict parameter setting and long computation time, three mountain factors are updated dynamically in IIGA to enhance the balance between local exploitation and global exploration. Consequently, the electrical connection of the TEG system is dynamically adjusted and reconfigured via IIGA, which can effectively improve power generation efficiency and reduce power consumption. The effectiveness and reliability of IIGA are testified and compared against four wellestablished meta-heuristic algorithms under two cases, i.e., 9 x 9 small TEG system and 15 x 15 large TEG system. Simulation results indicate that the output power boost by IIGA reaches 7.502 % under the 9 x 9 small TEG system and 10.281 % under the 15 x 15 large TEG system. Furthermore, a hardware-in-the-loop (HIL) experiment based on RTLAB platform is undertaken to testify the hardware implementation feasibility of IIGA affected by NTD conditions.