High Resolution PV Power Modeling for Distribution Circuit Analysis

Abstract

ii AcknowledgementsThe author wishes to express his appreciation to David Chalmers for implementing the 1-minute SolarAnywhere resolution, and to Phil Gruenhagen for creating the user interface to build systems and access the simulation service. 3 Executive SummaryNREL has contracted with Clean Power Research to provide 1-minute simulation datasets of PV systems located at three high penetration distribution feeders in the service territory of Southern California Edison (SCE): Porterville, Palmdale, and Fontana, California. The resulting PV simulations will be used to separately model the electrical circuits to determine the impacts of PV on circuit operations.The 1-minute simulations incorporate satellite-derived irradiance data with a spatial resolution of nominally 1 km x 1 km and a temporal resolution of 30 minutes. The special resolution is the highest available through existing satellite imagery, and is shown in Figure ES-1 for the Porterville site relative to the size of the modeled PV system.To obtain the 1-minute data, inter-image interpolations are generated with a "cloud motion vector" method by translating the previous image over time using wind speed and direction. The resulting irradiance data is fed into a PV simulation model to produce power output. As shown in Figure ES-3, the number of significant ramping events is very small, but the magnitude of the highest events is significant. The number of ramping events higher than 50% of PV system rated output per minute is taken as a metric of "significant" ramping, and this is shown for the Fontana site to be 37 events per year. The highest such event is shown in Figure ES-4 with an increase in PV output (caused by a departing cloud) of 2.20 MW per minute, or 75% of the systems rated output.Through methods such as the one described in this report and demonstrated through the datasets delivered under this project, utility engineers will be able to better predict the impacts of high penetration PV on their distribution circuits. List of Figures BackgroundSolarAnywhere is the premier solar irradiance time-series source for all locations within the continental US, Hawaii, Mexico, the Caribbean and parts of Canada. Irradiance estimates are generated using NOAA GOES visible satellite images processed using the most current algorithms developed by Dr. Richard Perez at the University at Albany (SUNY). These algorithms extract cloud indices from the satellite's visible channel using a self-calibrating feedback process that is capable of adjusting for arbitrary ground surfaces. The cloud indices are used to modulate physically-based radiative transfer models describing localized clear sky climatology. SolarAnywhere irradiance estimates have several advantages over ground based measurements including longer histories, lower costs, faster time to market, and the ability to directly produce solar power and variability forecasts.Clean Power Research works with Dr. Perez and SUNY to capture the latest advances in methodology and improvements to consistently pr...