Partial shadowing of photovoltaic arrays with different system configurations: literature review and field test results

Partial shadowing has been identified as a main cause for reducing energy yield of grid-connected photovoltaic systems. The impact of the applied system configuration on the energy yield of partially shadowed arrays has been widely discussed. Nevertheless, there is still much confusion especially regarding the optimal grade of modularity for such systems. A 5-kWp photovoltaic system was installed at K.U. Leuven. The system consists of three independent subsystems: central inverter, string inverter, and a number of AC modules. Throughout the year, parts of the photovoltaic array are shadowed by vegetation and other surrounding obstacles. The dimensions of shadowing obstacles were recorded and the expectable shadowing losses were estimated by applying different approaches. Based on the results of almost 2 years of analytical monitoring, the photovoltaic system is assessed with regard to shadowing losses and their dependence on the chosen system configuration. The results indicate that with obstacles of irregular shape being close to the photovoltaic array, simulation estimates the shadowing losses rather imprecise. At array positions mainly suffering from a reduction of the visible horizon by obstacles far away from the photovoltaic array, a simulation returns good results. Significant differences regarding shadow tolerance of different inverter types or overproportional losses with long module strings could not be confirmed for the system under examination. The negative impact of partial shadowing on the array performance should not be underestimated, but it affects modular systems as well as central inverter systems. (C) 2003 Elsevier Ltd. All rights reserved.