Project Background

The power grid faces daily challenges to keep our community powered. One of the primary issues is producing the right amount of energy to meet demand. Since there is limited storage, it is important to supply as close to the demanded power as possible. Otherwise, the power company risks under-generating thus some community members would lose power or over-generating thus wasting energy or potentially damaging generation and consumer equipment. If the power company is surprised by a sudden demand, they often have to start up other more responsive, but also more expensive, power plants to compensate for the difference in supply and demand.

Thduckcurvee graph at left illustrates the importance of these forecasts for generation dispatch. It uses actual data from 2012 and 2013 and projections to 2020 for the net load on conventional generators in the state of California. (That is, the total demand minus the supply from solar and wind generation) As more solar generators penetrate the market, the net demand sees a decrease to the middle of the day, followed by a sharp increase as the sun sets. At the scale that is projected to be reached within several years, the depth of this trough and the rate of change can vary by several Gigawatts depending on the output of the solar generators. A forecast of one to two hours ahead for a large number of the solar generators would reduce a large uncertainty in the problem of power dispatch.

Our senior design project addresses these problems by exploring forecasting of solar panel power output. As more renewable energy sources are being added to the grid, there is an increasing amount of uncertainty in power supply (and consequently power demand). This becomes even more complicated as both utility companies and consumers are incorporating photovoltaic systems. Therefore, both utility companies and Independent System Operators (ISOs) are interested in forecasts of power production. We are focusing on short-term (one to two hour) forecasts which is important for real-time power generation and storage adjustments.

-
Makai Mann
ESE Undergraduate
makai.mann@wustl.edu
-
David Sehloff
ESE Undergraduate
dsehloff@wustl.edu
-
Dr. Arye Nehorai
ESE Department Chair and Professor
nehorai@wustl.edu
-
Alex Cassidy
ESE PhD Candidate
cassidy@ese.wustl.edu

Contact Form: