PV-based Li-ion battery charger with neural MPPT for autonomous sea vehicles

In this paper a photovoltaic (PV) battery charger based on a DC-DC boost converter for a small size marine autonomous vehicle (AUV) is developed. The proposed solution employs a neural-based technique to estimate the solar irradiance on the basis of the actual PV panel voltage and current. This information is then used to perform an effective maximum power point tracking (MPPT) to optimise the energy exploitation of the solar panel. In particular the growing Neural Gas Network is used. The design and set up of the PV charger is presented together with experimental results assessing its performance.