SBIR-STTR Award

Goals of 80% clean energy production for the United States by 2035 and 20% of the countrys power being supplied by wind energy by 2030 imply nearly a tenfold increase in wind power production. This means that the need for forecast information will extend to longer projection windows with increasing penetration of wind power into the grid and also with diminishing reserve margins to meet peak loads during significant weather events. In addition to more complex issues regarding maintenance planning, energy trading of oil and gas will be influenced increasingly by anticipation of wind generation on timescales of weeks to months, and on longer time scales, future scenarios on decadal time scales are needed to support assessment of wind farm siting, government planning, and the regulatory environment. CFAN will expand upon its hybrid statistical/dynamical forecasting scheme that delivers probabilistic wind forecasts on time scales from one day to seven months to deliver ensemble-based forecasts and extended range outlooks that corrects not only for model bias error, but improves the shape of the distribution to capture low wind and high wind events that are critical forecast targets for wind farms. Collectively, the Phase I evaluation and testing of forecasts on different time scales utilizing our techniques will provide an assessment of the ability to overcome the existing challenges with wind power forecasts on time scales of days to months, as well as provide a framework to improve placement of future wind farm locations. The resulting efforts of a full Phase II project will deliver a wind energy decision support tool solution that would address three key market needs: the ability to forecast reliably at ranges beyond two weeks and at high spatial resolution; better capacity planning forecasts related to extreme or ramp events on scales of days to weeks; effective planning forecasts at decadal time scales that translate accurately for localized sites. The short-term market potential of a resulting Phase II solution is estimated to be in excess of $5M.

This proposal addresses the challenge of providing weather and climate information to support the operation, management and planning for wind-energy systems. There is a growing need for extended range forecast information as wind power increases its penetration into the grid. Future scenarios on decadal time scales are needed to support assessment of wind farm siting, long-term purchase agreements and the regulatory environment. To address this challenge, CFAN has developed a hybrid statistical/dynamical forecasting scheme for delivering probabilistic forecasts on time scales from one day to seven months using what is arguably the best forecasting system in the world (European Centre for Medium Range Weather Forecasting, ECMWF). The project also provides a framework to assess future wind power through developing scenarios of inter-annual to decadal climate variability and change. The Phase I project conducted a pilot study for the Electric Reliability Council of Texas (ERCOT) region. The project included: assembly and evaluation of relevant data sets; development and evaluation of an ensemble prediction framework for forecasting regional wind power generation and demand on time scales from days to months; development of strategies to assess long term (decadal) changes to the regional wind power environment; and formulation of an online tool that provides decision makers with actionable information related to wind power forecasts and projections. The proposed Phase II project will further develop the capabilities begun during Phase II and extend the project to include the continental U.S. and offshore regions. The objective is to develop a commercially viable capability for the growing array of diversified users in the wind energy forecast market. Commercial Applications and Other

Benefits:
Customers and end-users of these products include wind farm operators, regional power providers, grid system operators, and energy sales and trading. The other target application is assessment of wind energy project feasibility, to select favorable sites where wind is strongest and year-to-year variability is minimized. Customers for such assessments are project owners, government planners, regulatory agencies, and investors.