A method for error compensation of modeled annual energy production estimates introduced by intra-hour irradiance variability at PV power plants with a high DC to AC ratio
Daniel R Cormode, Nate Croft, Scott Kottmer, Rachel Hamilton
Bay4 Energy Services, Tucson, AZ, United States

Financial models of PV power plants, created during the development phase of a project are typically based upon simulations performed at hourly timesteps. Historical irradiance data used in these simulations, such as a Typical Meteorological Year, are created from average irradiances over hour-long periods. We find that when actual weather conditions include significant intra-hour variability in irradiance, the annual energy production estimates based on modeling in hour timesteps will not account for all clipping which occurs during moments of high irradiance. This is particularly significant for systems with high a DC to AC nameplate ratio. At some PV plants already in operation, this error has been found to result in overestimated production estimates of nearly 5%. A methodology is developed for compensating for this error in post processing, through consideration of observed intra-hour variability at geographically similar locations and applying a variable discount to each hour of the year. The discount function is intended to account for the difference between useful insolation as computed from hourly average irradiance versus a higher sampling rate, such as five minute average irradiance.