UC San Diego

As solar energy penetration in the power grid increases, challenges arise from the variable nature of solar power. Like any distributed generation source or load, photovoltaic (PV) generation can impact the voltage profile of a distribution feeder, potentially driving the service voltage outside of acceptable ranges. However, distributed energy resources can also bring great opportunities to improve power quality and cost effectiveness of the current power grid. In this dissertation, we will examine different strategies to increase solar PV penetration while complying with voltage standards and maximizing economic benefits of PV systems. The dissertation is divided into three main areas. The first one covers our solar forecasting technology using sky imagery and proposes an improvement to the current cloud height calculation using a stereographic method with multiple sky imagers. These sky images are then used in the second part to assess the impact of solar variability on distribution feeders with high solar PV penetration. Quasi steady-state power flow simulations at different solar penetration levels and cloud conditions are forced by site-specific solar resource data obtained from high resolution sky imagery. The impacts of variable solar power on voltage fluctuations, high frequency and magnitude power ramps, and transformer tap operations will be considered. The final area investigates how smart solar PV inverters can resolve voltage issues on distribution feeders. Once reactive power injection by smart inverters is allowed, overvoltages during midday and undervoltages during the evening can be avoided given a sufficient number of smart inverters on the distribution feeder. An analysis on generic feeders is followed up with a field test that demonstrates the potential benefits of smart inverters.