An artificial neural network-based real time maximum power tracking controller for connecting a PV system to the grid

This work deals with the application of a neural network-based controller for tracking the point of maximum power of a photovoltaic (PV) system interconnected to the utility grid. The neural network is used to identify, in real time, the voltage for maximum output power of the system. The controller, through the information supplied by the neural network, generates a control signal that will be applied to a DC/DC (boost) converter in such a way to take the voltage of the system to a value which guarantees the operation of the PV system at maximum power. The boost converter duty-cycle is generated by a PI controller based on the information supplied by the neural network. In order to connect the PV system to the electric distribution system a three-phase voltage source inverter (VSI) is used operating with optimized sinusoidal PWM strategy with harmonics elimination at the output voltage up to the 17/sup th/ harmonic. The inverter uses IGBT as power switches, and is microcontroller operated.