Research on impacts of distributed versus centralized solar resource on distribution network using power system simulation and solar now-casting with sky imager

As solar energy penetration in the power grid increases, utilities face great challenges coping with the variable output of solar power, especially from photovoltaics (PV) panels. In this research, we investigate the impact of distributed versus centralized solar resource on distribution network using power system simulation and solar now-casting with sky imager. The distribution network studied is a large and realistic distribution system in San Diego area modelled by our research team at UCSD and validated against power flow and short circuit data given from a local utility. The network has 1733 customer loads, summing up to 11.1 MW demand, and 2.3 MW PV in peak power production. Special scenarios are set up to simulate the difference between distributed and centralized solar resource with PV penetration levels (defined as max PV power output divided by max load demand) from 0 to 100%. OpenDSS is used for quasi steady-state simulations of the network on a partly cloudy day (December 12th, 2012). Solar now-casting using UCSD Sky Imaging (USI) system is utilized to provide unique PV production curves for all PV systems on the network. The impacts of solar resource including voltage level, line losses, power consumption and number of tap operations are reported for each investigated case and. It is shown that centralized configuration is much more severely impacted by high PV penetration than distributed set up in terms of high voltage level, increasing number of tap operations, and reverse power flow.