Trajectory2

Optimizing Lidar Design for Wind Turbine Control Applications

Date: July 5th 2016

Venue: Boston Marriott Copley Place, Boston, MA, USA

Immediately preceding the 2016 American Control Conference

Workshop leader: Eric Simley, Envision Energy

DSC_9613

Summary

Lidar-assisted control using preview wind information has become an important research topic in the wind turbine control community, and several lidar manufacturers offer forward-staring nacelle lidars that can be used for control purposes. The use of lidar-assisted control is currently limited to a few research turbines, but results are promising. For example, during initial field testing collective pitch feedforward controllers have been able to successfully improve rotor speed regulation, reduce fatigue loads, and decrease pitch actuator activity. Additionally, numerous control studies relying on simulation are paving the way for more advanced lidar-assisted controllers. One obstacle to the widespread adoption of lidar-assisted control, however, is its multi- and interdisciplinary character. A thorough understanding of lidar measurement principles and limitations (e.g., reliability, accuracy, scan pattern capabilities, etc.) is mandatory for designing a control system. Also, knowledge about wind turbine dynamics and controls is necessary to determine what types of lidar signals are useful for control and what lidar system properties are required. Since lidar and turbine manufacturers typically specialize in their own part of the puzzle, current lidar systems are not necessarily optimized for wind turbine control applications. The purpose of this workshop was to bridge the gap between lidar manufacturers and wind turbine control engineers by identifying the objectives of lidar-assisted control, barriers preventing the widespread use of lidars for control, design suggestions for overcoming those barriers, and additional design considerations for maximizing the effectiveness of lidars for control purposes. The workshop began with an invited presentation giving an overview of lidar-assisted control, followed by invited presentations from wind turbine manufacturers and lidar suppliers. The lidar manufacturers addressed the objectives of lidar-assisted control as well as the barriers, while speakers from the lidar industry examined the capabilities of current lidar systems and strategies for improving lidar technology for control applications. Afterwards, participants gathered in small groups to define the objectives of lidar-assisted control and identify the barriers preventing greater use of lidar-assisted control. The workshop continued in the afternoon with an overview of sources of lidar measurement error for control purposes and how lidar scan pattern design choices affect measurement quality. To familiarize all participants with some theoretical aspects of lidar optimization and controller design, workshop attendees then participated in two MATLAB exercises, which were sent out prior to the workshop. The first exercise examined how to minimize measurement error by optimizing a simple lidar measurement scenario and the second exercise explored how to most effectively use those measurements in a simple control system. The program concluded with another round of small group discussions where all participants were able to share their perspectives on how lidars can be optimized for control applications, specifically addressing the barriers that had been identified. As a follow-up to the workshop, we are preparing a tutorial paper on optimizing lidars for control applications.

Public Documents

  • Invitation (pdf)
  • Details (pdf)
  • Minutes (pdf)

Internal Documents

This section with the presentations from the workshop is password protected. The password will be send out to all task participants. If you missed the email, please contact us.

Photos

  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image
  • Demo Image