Power ramp rates and variability of individual and aggregate photovoltaic systems using measured production data at the municipal scale

Intermittency from photovoltaic systems can negatively impact electricity grid stability due to increased power ramp rates and temporal misalignment between generation and load. Distributed systems, especially rooftop residential systems, present a range of orientations and spatial separation that when aggregated introduce temporal diversity. If these features reduce negative impacts, they should be weighed against increased costs per delivered energy compared to larger commercial/industrial systems. We compare power production data from 60 distributed residential PV systems spread across a large municipality with medium and large commercial rooftop systems located centrally in the municipality. Power ramp rates and power output variability were calculated. The aggregated distributed residential system ramp rates never exceed 13% per 5 min, while the medium/large commercial systems experienced ramp rates up to 61% and 68% per 5 min. In aggregate the residential output variability is 3 times lower than the commercial sites. Consistent with previous literature the output variability decreases as the number of systems grows. There are however diminishing returns due to the limited geographic area, with a relative output variability for the municipality approaching 20% of the variability of a single system. Residential data was scaled up to represent 10% of electrical load and resulted in 10% more electricity grid ramp rates exceeding ±5 MW per 5 min compared to baseline. Finally, a comparison between pyranometer and photovoltaic system ramps rates shows the latter are less severe, likely because pyranometers are single points sources.