The objective of this project is to develop and apply experimental and computational tools for predicting and improving wind farm performance by placing particular attention on large scales of turbulence and vertical fluxes of kinetic energy that are of great significance for large arrays of wind turbines. Much effort has been devoted in recent years to increasing the efficiency of individual wind turbines, assuming a given inflow in front of the turbine. Also, understanding how wakes affect the performance of downstream turbines and modeling superpositions of multiple such wakes has received considerable attention; however, there has been relatively little fundamental understanding of how a large array of wind turbines interacts with the turbulent atmospheric boundary layer at larger scales in the wind turbine array boundary layer (WTABL). Recent research has demonstrated that an important performance-limiting factor for large wind farms is the rate at which kinetic energy can be entrained into the array from the flow aloft, above the wind turbines. No matter how efficient an individual wind turbine is, or how well it can adapt to an upstream wind turbine, ultimately it is the vertical flux of kinetic energy into the overall array that largely determines how much power can be extracted from the atmospheric flow. The questions addressed in this project aim at better understanding the limiting factors and the effects of different scales of turbulence on vertical entrainment processes. The resulting models should guide wind turbine placement strategies and possible flow modifications so that vertical entrainment rates can be increased. Specifically, wind tunnel experiments coupled with large-eddy simulations (LES) will be employed to address the following research questions: (a) What are the essential differences between the developing and the fully developed WTABL? (b) What is the relative contribution from streamwise large-scale coherent vortices to vertical entrainment of kinetic energy? (c) What are the space-time correlations of hub-height velocity and power output between different wind turbines in the array? (d) Are there particular arrangements of wind turbines in the array that increase, on average, the entrainment? and (e) Can large-scale flow structures be affected through rotor modifications to increase such entrainment? Addressing such questions requires the ability to experiment under the highly controlled and reproducible conditions that can be afforded in the wind tunnel experiments and computer simulations. The data will be supplemented with comparisons with relevant new field data from a large wind farm.

Broader impacts: The robust growth of wind energy implies the possibility that large portions of the land and near-shore surface of the US and the world may ultimately be used for large wind farms. Predicting and better understanding the physical processes coupling the modified surface and atmosphere under such conditions is a timely and critical area of research. Through project activities the PIs will help train the next generation of engineers and scientists with the necessary tools and insights to help reach the US goal of 20% wind energy by 2030. Graduate education/mentoring will stress the interplay between wind tunnel experimentation, computer simulation and field data analysis. International (Switzerland, Spain) and industrial experiences (General Electric) will also be emphasized in this project. Recruiting and outreach will leverage both PIs' ongoing efforts to recruit US Hispanic graduate students through contacts in Puerto Rico (NSF-AGEP and LSAMP), as well as an IGERT at JHU on modeling complex systems. A GK-12 at RPI on energy and environment will leverage NSF resources in training teachers on wind energy issues. The PI's ongoing outreach to a Baltimore high school will be continued, providing research experiences for high-school juniors and seniors.

Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Johns Hopkins University
United States
Zip Code