The objective of this research is to develop the concept of Smart Solar Farms. The approach is to integrate advanced solar forecasting methods into parallel computer platforms with the objective of quantifying the variability and availability of solar energy for power production, thus enabling solar integration with the power grid.

Intellectual Merit This project integrates a very high number of inputs, varying from radiometric and meteorological time series to high-resolution remote sensing and ground-to-sky images, into a single adaptive and intelligent learning platform capable of local processing of high-fidelity forecasts. The proposed research is transformative in that novel, highly optimized natively parallel platforms will generate, in-situ, real-time forecasts for all time horizons of interest by a combination of machine learning software/hardware integration with intelligent energy management systems.

Broader Impacts The reliable and economical integration of solar energy resources into the power grid is one of the most critical global technological challenges of our time. This project develops cost-effective tools that are essential to the integration of solar energy into a variety of possible smart grid solutions that aims at energy security and power quality improvements. We will also educate a new diverse generation of green energy experts, who master critical machine learning techniques specifically designed for variable resource power production and control. The concept of Smart Solar Farms will be demonstrated, and the test-bed will continue to provide high quality information, research opportunities and educational experiences for the community-at-large for many years beyond the duration of the project.

Project Start
Project End
Budget Start
2012-05-01
Budget End
2015-04-30
Support Year
Fiscal Year
2012
Total Cost
$384,452
Indirect Cost
Name
University of California San Diego
Department
Type
DUNS #
City
La Jolla
State
CA
Country
United States
Zip Code
92093