Skip to content Skip to navigation

A Commercial Wind-Farm Testbed for Performance-Prediction and Optimization Tools


photo: Barry cropped from Flickr

John O. Dabiri, Civil and Environmental Engineering and Mechanical Engineering; Sanjiva Lele, Aeronautics, Astronautics, and Mechanical Engineering

The U.S. Department of Energy estimates that, on average, wind farms have produced a mere 30% of their potential output capacity since 1999. This means that wind-farm operators could be generating more renewable energy and realizing more profits. To increase energy production and minimize operation costs for these operators, John Dabiri and Sanjiva Lele have each been developing wind-farm performance-prediction and optimization tools. This joint project will allow them to demonstrate the combined effectiveness of some of these tools. The Firewheel Wind Energy Complex currently under development in Texas and Oklahoma will serve as the primary testbed for this project. The facility, which may become the largest wind farm in the world, provides a unique opportunity to coordinate the placement of meteorological sensors and aerodynamic data collection before, during, and after site construction. An improved ability to predict the energy production of the wind farm will reduce economic and logistical uncertainty associated with this and future wind farms.

The key technological advances that will be used at the testbed will include low-order modeling of complex, multiscale flow physics. Specifically, the researchers will use tools they have each developed to accurately simulate the interaction of wind turbines with the atmospheric boundary layer, on scales ranging from individual wind turbines to large turbine arrays in farms. The low computational cost of the models will enable the evaluation of hundreds of potential wind-farm configurations in less than an hour on a single CPU, thereby facilitating optimization on time scales relevant to commercial project development. This turbine-specific computational modeling is complemented by a new framework for kinematic simulation of the atmospheric boundary layer. Using focused analyses of the length and time scales of the wind that are most relevant to the wind farm, energy transport within the wind farm can be computed with up to four orders of magnitude less computational expense than current modeling techniques.

During this project, the turbine-specific models will be extended from their previous application to vertical-axis wind turbines, simulating instead the horizontal-axis wind turbines at Firewheel. The high spatial and temporal resolution of the wind and turbine power measurements collected during the phased construction of the wind farm will provide for a much more rigorous validation of these computational tools than what is currently available. Once the efficacy of the wind farm simulation tools has been characterized and refined based on measurements from the initial phase of turbine installations, the same tools will be used to optimize the remainder of the wind farm layout. This multi-objective optimization will include considerations for energy production, fatigue loading, construction costs, and logistics of operations and maintenance.

 

Awarded 2016