Fast atom search optimization based MPPT design of centralized thermoelectric generation system under heterogeneous temperature difference

Abstract

This paper designs a novel maximum power point tracking (MPPT) technique for centralized thermoelectric generation (TEG) system under heterogeneous temperature difference (HeTD). Since the HeTD can result in multiple local maximum power points (LMPPs) for centralized TEG system, a fast atom search optimization (FASO) is designed to approximate the global maximum power point (GMPP) from multiple LMPPs. In order to accelerate the convergence for a high quality solution, the Euclidian distance ratio of the original atom search optimization (ASO) is employed to adaptively update according to the dynamic optimization results, as well as for the number of neighbours for each atom. Through case studies such as start-up test, step variation of temperature, stochastic temperature change, and analyse of sensitivity are carried out, in which the practicability and superiority of FASO are compared with that of MPPT method based on a single LMPP and four meta-heuristic algorithms. Simulation results show that the energy output produced by FASO ranges from 101.64% to 451.49% to the alternatives. Finally, through hardware-in-the-loop (HIL) experiment based on a dSpace, the feasibility of the hardware realization is confirmed, in which the difference between simulation results and HIL experiment results is less than 2.3%.