Maximizing Solar Forecast-Based Distributed energy Resources

Solar photovoltaic (PV) variability and uncertainty limit solar penetration into the electric power system. Solar forecasting helps reduce the uncertainty of solar PV. The University of California, San Diego research team developed potential “use cases" for distributed energy resource technologies that include solar PV, solar forecasting, and controllable loads. The distributed energy resources can provide benefits to the electric grid and the distributed energy resource-owning utility customer.
The research team used nine sky imaging cameras in the Greater Los Angeles area with a specific focus on warehouse rooftop areas. More accurate forecasts due to real-time updates from these cameras enabled smarter charging of a fleet of electric vehicles at the workplace. Electric vehicle fleets range in number and vehicle sizes, up to a school bus. PV forecasts are leveraged to shape a combined electric vehicle load profile that consumes excess PV generation. The smart charging algorithm was successful in flattening the net load. This study also considered the economics of selected use cases. Forecasting-aware scheduling benefits the customer the most by reducing demand charges. Conversely, wholesale market sales of PV energy directly to the California Independent System Operator (California ISO) day-ahead market are economically unattractive. The ratepayer benefit of reducing loads during high grid stress are small due to high reliability of the California ISO system. Recommendations are provided to increase the availability of daytime electric vehicle charging to support the concepts developed in this contract.