In this paper a maximum power point tracking algorithm able to drive the operating point of a partially shaded photovoltaic system toward the global maximum power point is analyzed. The algorithm exploits detailed information about the electrical parameters of all solar panels forming the solar system gained by means of a distributed sensor network, which monitors the operation of the solar field at a very high granularity level. Data collected by the monitoring system are exploited to reconstruct the power voltage curve of the photovoltaic system, thus recognizing the presence of multiple local maxima and their exact voltage position.
Experiments performed on a pilot solar filed equipped with the sensor networkevidence the reliability of the analyzed approach. A convergence time of about 2.5 s was achieved independently of illumination conditions and, in case of partial shadowing, an increment of 90 W (50% more) with respect to a standard tracking algorithm.