Neural Control for Photovoltaic Panel Maximum Power Point Tracking

With the rise in the use of renewable energies, solar panels have proven to be reliable and have a favorable cost-benefit ratio, producing energy free of noise and air pollution. Solar panels are subject to considerable variations in working conditions due to changes in solar irradiation levels and temperature that affect its semiconductor properties. To be able to profit as much as possible from this source of energy, control of the modules and perturbation rejection is very important to obtain the highest viable amount of electrical power. This work is concerned with the on-line identification and control of a photovoltaic system using neural networks with the Kalman Filter as training algorithm. Having on-line identification and control allows the system to be more adaptable to changes in weather and other variations than with common off-line methods.