An Enhanced Maximum Power Point Tracking Method for Thermoelectric Generator Using Adaptive Neuro-Fuzzy Inference System

The design and performance study of a variable fractional factor combined adaptive neuro-fuzzy inference system (VFANFIS) based maximum power point tracking (MPPT) technique for the thermoelectric generator (TEG) is presented. In this proposed MPPT technique, the fractional factor is used in such a way to achieve the variable tracking step size based on the operating point of the TEG in the power versus voltage (P–V) curve. The larger tracking step size move the TEG system operating point quickly towards the maximum power point (MPP), suppose the TEG operating point reaches near to MPP then the tracking step size becomes smaller to maintain stable output without any oscillation. In the proposed algorithm, the fractional factor value which is determined based on the change in voltage of the TEG is used expand or contract the input domain range of the ANFIS by which the variable tracking step size is achieved. The energy efficiency of the TEG can be improved by attaining the MPP quickly and maintaining steady-state output around the MPP. The effectiveness of the VFANFIS-based MPPT technique is demonstrated with MATLAB simulation study under different thermal and electrical operating condition. The study results are confirmed that the proposed MPP tracker performs better to extract the maximum power from the TEG by achieving the MPP quickly and accurately when compared to the conventional incremental conduction (INC) based MPPT technique.