An experimental procedure to check the performance degradation on-site in grid-connected photovoltaic systems

A power and energy procedure in operating photovoltaic systems to quantify the losses according to the causes

“…It is interesting the inverse relationship of fill factor with power deviations vs. datasheet: low values of FF (63-67%) correspond to high deviations vs. datasheet [6].…”

“…Then, as a first experimental test, it is recommended to distinguish very quickly portions (arrays) of the PV field, in which no further experimental tests are required (acceptable performance) or, on the contrary, deep analysis is required at string and PV module levels. This ability, which minimizes the lost production during the disconnection of PV modules to carry out their tests (<1 s per each test), is given by the I-V curve measurement at level of PV array without the typical size limits of the commercial instrumentations, i.e., 100 kWp as written in [5], [6]. Moreover, the uncertainty of the measurements is guaranteed by repetitive calibrations of the Automatic Data Acquisition System (ADAS) used in the on-site testing.…”

“…The automatic data acquisition system, described in [6], is periodically calibrated and is made of: The I-V curves of a PV array was measured with a capacitive load, detecting simultaneously voltage, current, irradiance and temperature [16], [17]. The next data-processing to achieve the Standard Test Conditions (STC, G=1 kW/m 2 , T c =25 °C) parameters is performed by a conventional correction method [18].…”

“… for the irradiance G the absolute uncertainty is ±20 W/m 2 , for the ambient temperature T a the absolute uncertainty is ± 0.2 °C, for the cell temperature T C the absolute uncertainty is ± 2 °C (indirect measurement according to [6], [19] whit open circuit voltage and its thermal coefficient);…”

“…17). A careful examination of the data reported in [6] permit clarifying the correlation between power loss and cracking in modules affected by the simultaneous presence of PID and cracks. In general, the two phenomena have a similar impact regarding the power losses and therefore a clear trend between number of observed cells and the power losses cannot be deduced.…”