Photovoltaic devices used as energy sources exhibit a non-linear and time-varying current-voltage characteristic. Power converters ensure that the operating point of the photovoltaic source is always near maximum power. In this work, a Maximum Power Point Tracking algorithm variation is proposed. This approach uses an algorithm to construct a dynamic dataset, which in turn is used to train a neural predictor that aides the algorithm initialization. The approach is validated in a simulated environment considering a SEPIC converter at steady state, including the parasitic components. The algorithm performance is compared against classic Incremental Conductance showing the obtained advantages in terms of quicker convergence and adaptability to source degradation over time.
Lozito, G.M., Grasso, E., Fulginei, F.R. (2022). A Neural-Enhanced Incremental Conductance MPPT Algorithm with Online Adjustment Capabilities. In 2022 IEEE International Conference on Environment and Electrical Engineering and 2022 IEEE Industrial and Commercial Power Systems Europe, EEEIC / I and CPS Europe 2022 (pp.1-6). 345 E 47TH ST, NEW YORK, NY 10017 USA : Institute of Electrical and Electronics Engineers Inc. [10.1109/EEEIC/ICPSEurope54979.2022.9854781].