PV plants require a monitorization of their electrical parameters to achieve its best performance. Lately, it is more and more common that SCADA systems monitor currents and voltages even of every single string. These measurements which use to be stored instantaneously with a frequency like the one established on inverters and energy meters (usually each 10, 15, 20 or 60 minutes) are useful to detect faults. Unfortunately, it is not taken advantage of them to characterize the PV arrays. A better option would be to exploit this monitoring capability not only to detect operation failures, but also to characterize the PV strings/arrays. In order to implement this characterization it is needed to use the adequate devices to measure voltages and currents and determine which should be the optimal frequency to store data, depending on they are instantaneous or mean values. This paper presents the results of the DC monitoring of a grid-connected PV array of 5.8 kWp, installed on the headquarters of the Instituto de Energía Solar of the Universidad Politécnica de Madrid (IES-UPM), located in the Campus Sur of Vallecas (Madrid), which allows characterizing the PV arrays power. To study the relation between sampling frequency and its impact in the accuracy of the power averages each 1’, 5’, 10, and 15’ have been calculated from instantaneous measurements, and they have been analysed to evaluate which is the more adequate sampling rate for instantaneous and mean values.

31st European Photovoltaic Solar Energy Conference and Exhibition; 2366-2369