oa Modelling The Power Produced By Photovoltaic Systems

The development and improvement of a model that can provide accurate estimates of the power produced by a photovoltaic system is useful for several reasons. A reliable model contributes to the proper operation of a photovoltaic power system since any deviations between modeled and experimental power can be flagged and studied for possible problems that can be identified and addressed. It is also useful to grid operators to know hours or a day ahead the contribution from different PV systems or renewable energy systems in general. In this way, they will be able to manage and balance production and demand. The model was designed to use the smallest number of free parameters. Apart from the incoming irradiance and module temperature, the model takes into account the effects introduced by the instantaneous angle of incidence and the air mass. The air mass is related to the position of the sun during its apparent motion across the sky since light travels through an increasing amount of atmosphere as the sun gets lower in the sky. In addition, the model takes into account the reduction in efficiency at low solar irradiance conditions. The model is versatile and can incorporate a fixed or variable percentage for the losses due to the deviation of MPPT tracking from ideal, the losses due to the mismatch of the modules, soiling, aging, wiring losses and the deviation from the nameplate rating. Angle of incidence effects were studied experimentally around solar noon by rotating the PV module at predetermined positions and recording all necessary variables (beam & global irradiances, module temperature and short circuit current, sun and module coordinates). Air mass effects were studied from sunrise to solar noon with the PV module always normal to the solar rays (global irradiance, temperature and short circuit current were recorded). Stainless steel meshes were used to artificially reduce the level of the incoming solar irradiance. A pyranometer and a reference cell were placed behind the mesh, while the unobstructed solar irradiance was monitored with a second reference cell. The different mesh combinations allowed us to reach quite low levels of irradiance (10%) with respect to the unobstructed irradiance (100%). Seasonal dust effects were studied by comparing the transmittance of glass samples exposed to outdoor conditions, at weekly time intervals, against a cleaned one. Data from several different US sites as well as from PV systems located in Crete, Greece are currently used to validate the model. Instantaneous values as well as daily integrals are compared to check the performance of the model. At this stage of analysis, it turns out that the typical accuracy of the model is better than 10% for angles of incidence less than sixty degrees. In addition, the performance of the model as a function of the various parameters is being studied and how these affect the calculations. In addition to the functions that have been determined from our measurements, functions available in the literature are also being tested.