Improving Solar & Load Forecasts: Reducing the Operational Uncertainty Behind the Duck Chart
Improvement of net-load forecasting to reduce scheduling errors for the California grid.
Itron, Inc., dba IBS
Recipient
Davis, CA
Recipient Location
3rd
Senate District
4th
Assembly District
$925,538
Amount Spent
Completed
Project Status
Project Result
The project incorporated several forecast improvements by advancing methods for determining BTM system specifications and shading based on measured production inputs, integrating irradiance measurements to improve aerosol optical depth and cloud albedo aspects, and by incorporating near real-time metered PV generation data to fine-tune fleet forecasts of both grid-connected and BTM PV solar. Researchers evaluated three alternative model approaches for extending the CAISO load forecast framework and present the alternative load forecast frameworks for incorporating BTM solar PV forecasts. The study showed that improvements in solar and net load forecasting can provide positive financial impacts in the scheduling and procurement of electricity in the wholesale electric market within the State. The potential savings would have been on the order of $9 million just in the covered period.
The Issue
Solar is an intermittent resource, and accurate prediction of when and how this fluctuating renewable resource can be used is essential for grid operators. Increasingly accurate forecasting tools have been developed in recent years, but they have yet to be fully implemented into grid operations to optimize operations for high-penetration solar. Furthermore, none of the California ISO load forecast models include and capture the impact of behind-the-meter solar PV on measured loads. The limitation of integrating state-of-the-art solar forecasts into net-load forecasts is based on the absence of estimates to determine the value of utilizing improved PV solar forecasts into grid operations.
Project Innovation
This project aimed to reduce the operational uncertainty in both PV and net load forecasts by producing high accuracy forecasts and linking them to net load forecasts at finer time intervals. This increased accuracy in estimation and incorporation within net load forecasts will enable better integration of intermittent PV generation in California and lead to substantial savings in the associated wholesale energy market costs. The results of this agreement contribute to reduced operational uncertainty behind the Duck Chart by producing high accuracy solar generation forecasts for utilities and the CAISO, and linking these generation forecasts to methods for forecasting net loads at higher temporal resolution. This increased fidelity and connection to net load forecasts will provide critical insights to better manage the rapidly evolving grid in California.
Project Benefits
The project showed that improvements in solar and net load forecasting methods can provide positive financial impacts in the scheduling and procurement of electricity in the wholesale electric market within the State. The results of this research have shown that, just in the period covered by this analysis, the potential savings to all stakeholders would have been on the order of $9 million. With further growth in solar and improvements in integrating behind the meter solar into the California ISO net load forecasts, the team anticipates it can achieve even greater cost reductions. This research sets the groundwork for further research on developing a framework to optimize the use of alternative forecasts by the California ISO into its net load forecast. It may be possible to develop a framework for choosing when to use the alternative forecast to optimize its value to all stakeholders.
Affordability
Improved net load forecasts reduce the cost of grid regulation required to cover increasing load forecast errors. By reducing the percentage error by just 0.1 percent, the California ISO and California ratepayers can save more th
Environmental Sustainability
Reduced requirements for regulation services and spinning reserves will help reduce GHG emissions by an estimated 2.7 million tons per year through reduced use of natural gas fired peaker plants.
Reliability
The project increases system reliability by significantly increasing the accuracy of solar PV forecasts and the associated net load forecasts. This integration of state-of-the-art solar forecasts into net-load forecasts further e
Key Project Members
Stephan Barsun
Subrecipients
Clean Power Research
Match Partners
Clean Power Research
Itron, Inc., dba IBS
