Assessing the Ability of Smart Inverters and Smart Consumer Devices to Enable more Residential Solar Energy

Improved forecasting of solar irradiance in California, incorporating fog and stratus conditions.

Grid Decarbonization and Decentralization Variable Renewable Generation

Electric Power Research Institute, Inc.

Recipient

Palo Alto, CA

Recipient Location

13th

Senate District

23rd

Assembly District

beenhere

$746,321

Amount Spent

closed

Completed

Project Status

Download Project

Project Result

The project completed a measurement program involving targeted deployment of ground-based atmospheric boundary layer sensors. This successful measurement program can serve as a model for coordinating similar efforts in the future. Given the overlapping benefits of boundary layer sensor data across weather, air quality, and energy forecasting applications. The team performed many months of forecasts to investigate the impact of various physical parameterization choices on the WRF model. The results show an overall improvement in forecast skill and suggested that point measurements at a handful of locations are not likely to substantially improve forecasts above their already fairly high skill. The team also implemented and tested machine learning models for predicting cloudiness and solar irradiance at very short-term forecast horizons using project sensor data. Overall results could be integrated into operational tools of CAISO and utilities.

The Issue

Successful integration of renewable resources into power system operations will require the ability to forecast the output of these resources in timeframes from less than an hour to days ahead. Fog and stratus affect solar irradiance in California throughout the year, and shortcomings in predicting fog and stratus dissipation currently constrain the accuracy and confidence of short-term solar irradiance forecasts. The value of using improved forecasts is still not well understood by grid operators and utilities due to the difficulty of assessing return on investment for an improved forecasting system particularly for deploying instruments to improve the data used in forecasting models.

Project Innovation

The project develops an improved forecasting system for solar irradiance in California, with a particular focus on fog and stratus conditions, through targeted deployment of instrumentation. The improved forecasts will be integrated into operational tools for use by the California Independent System Operator (CAISO) and utilities. This project utilizes a targeted instrumentation network, consisting of existing and new sensors, to improve the models used for forecasting fog and stratus conditions. The Recipient will design and deploy this network with the aim of improving the forecasts that are most important to CAISO and utility operations.

Project Benefits

The use of an advanced network of existing and new instrumentation to inform numerical weather and statistical model improvements will significantly improve the current state of solar forecast modeling in California. The holistic forecasts produced will showcase a combination of various aspects of the weather forecast value chain, not previously demonstrated, linking observation systems and advanced physical and statistical modeling for solar forecasting. The project's focus on fog and marine layer forecasts, which are traditionally challenging to predict, will improve solar energy forecasting and contribute to increased PV penetration.

Affordability

Improved forecasts help reduce operating costs by improved commitment and dispatch of generating resources, reductions in solar power curtailment, and more optimal procurement of resources for Investor Owned Utilities (IOUs).

Economic Development

Improved forecasts of marine layer and fog conditions have the potential to improve the efficiency of generation dispatch, reduce the need for operating reserves to manage forecast error and maintain or increase reliability while integrating increased levels of renewables.

Reliability

Improved forecasts support the advancement of reliability of renewable energy by reducing uncertainties in generation across the CAISO system, improving voltage control on distribution systems, and ensuring that utilities can perform transmission and distribution switching.

Safety

Improved forecasts can help maintain safety at the distribution and transmission level, improve switching operations required for DER management, and inform new smart grid devices that can manage voltage.