Conference programme

Back to the programme printer.gif Print



Thursday, 19 November 2015
11:30 - 13:00 Wakes: LiDAR measurements, layout optimization and modelling uncertainties
Resource assessment  
Onshore      Offshore    


Room: Montparnasse

The focus of this session is on wind turbine wakes. Two studies deal with measurement of the wake wind field using LiDARs, and subsequently verifying them against reference instruments. One study deals with wind farm layout optimization in a complex terrain, where the wake wind field is generated using CFD tools. One study deals with providing a framework for evaluating wake models uncertainties, and subsequently their influence on the predicted wind turbine power in wakes.

Learning objectives

  • The importance of evaluating the uncertainty of the estimated wind speed in wakes using measurements from a Doppler LiDAR, with an example of how to do it under different atmospheric conditions
  • How a 3D wake wind field looks like from LiDAR measurements, potentially paving the way for its subsequent use in wake tracking and model parameterization
  • How CFD wake results in complex terrain can be combined with layout optimization techniques, potentially being applicable to wind farm design
  • A framework for wind farm flow model validation, where it is demonstrated how uncertainties in input variables propagate into resulting uncertainties in predicted wind turbine power in wakes
Lead Session Chair:
Ameya Sathe, DTU Wind Energy, Denmark
Hugo Herrmann EDF Energy R&D UK Centre, United Kingdom
Co-authors:
Hugo Herrmann (1) F Sami Barbouchi (1) Eric Dupont (2) Yannick Lefranc (2) Raghu Krishnamurthy (3) Matthieu Boquet (3) Jeremy Grenier (4) Cedric Dall Ozzo (4)
(1) EDF Energy R&D UK Centre, London, United Kingdom (2) EDF R&D, Paris, France (3) LEOSPHERE SAS, Orsay, France (4) EDF EN, Paris, France

Share this presentation on:

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2015 and are published here unedited

Mr. Herrmann is currently an offshore wind research engineer at the EDF Energy R&D UK Centre. He studied electrical power and computer engineering at the Ecole supérieure d’électricité in France before doing an MSc in fluid mechanics at Imperial College in London. His research is focused on wind resource assessment.

Abstract

Evaluation of wind speed uncertainty estimated with a scanning Doppler LiDAR inside and outside the wake of a turbine

Introduction

Part of the underperformance of large wind farms is attributed to an incorrect design of the layout due to a wrong estimation of the wake losses. Several wake models have been developed to estimate the wake behavior of a single turbine or multiple wind turbines. Scanning LiDARs can measure spatial variations of the wake and can be used to estimate the wind speed deficit, wake width and length of wind turbines up to 4-6 km away from the scanning LiDAR. The instantaneous wake structures measured by the scanning LiDAR provide great insight into the evolution of wake for various atmospheric conditions. Assessing the uncertainty of the scanning LiDAR measurements is crucial for validating various wake models. This study is focused on assessing the wind speed uncertainty from scanning LiDAR measurements.

Approach

A WINDCUBE 200S scanning LiDAR was deployed onshore to assess the accuracy and uncertainty of wake measurements inside and outside a wake compared to a traditional met-mast equipped with cup and sonic anemometers and also a co-located WINDCUBE V2 vertical profiler. Several scans from the scanning LiDAR were implemented towards the met-mast and turbine. Two types of scan patterns based on the wind direction were configured to assess the depth and length of wake measurements and accuracy of wind speed measurements at the location of the met-mast.

Main body of abstract

To assess the accuracy of wake parameters retrieved from a scanning Doppler LiDAR, an onshore deployment was conducted using a WINDCUBE 200S scanning Doppler LiDAR, WINDCUBE V2 vertical profiler and a traditional 80 m met-mast with cup and sonic anemometers. The wind turbines located in the vicinity, in certain wind directions, provided co-located wake measurements from all the three instruments. The wind speed measurements from all the three instruments were compared for data within a turbine wake and outside the turbine wake. The uncertainty of wind speed reconstructed from the scanning LiDAR for various heights were compared to met-mast measurements inside the wake, while outside the wake, measurements from both the met-mast and WINDCUBE V2 were compared. In this study, the scanning Lidar comparison results at various heights, scanning LiDAR reconstruction algorithm overview and the details of the scanning scenario from the scanning LiDAR along with setup challenges will be presented. The length and width of the wind turbine wakes are also assessed, and the statistics of wake behavior based on stability measurements (sonics) on the met-mast will be presented. In a second step, these measurements will be used to validate a computational Fluid Dynamics (CFD) tool “Code_Saturne” developed by EDF.

Conclusion

This study will focus on results from scanning LiDAR to met-mast (including sonics) & WINDCUBE V2 measurements. The uncertainty scheme as provided in the IEA Task 32 will be implemented to assess the accuracy of the scanning LiDAR wind speed estimates compared to standard met-mast or WINDCUBE V2 measurements as reference. The wind speed maps provided by the scanning LiDAR are great tools to assess the wind speed, the width and depth of the wake for comparison to CFD models. In the near-future, these measurements will be extended to assessing wake performance for an offshore campaign.


Learning objectives
1. The audience would get a clear picture of the scanning LiDAR installation protocols used for the campaign;
2. The uncertainty of scanning LiDAR measurements inside & outside the wake of a turbine;
3. The width & depth of wind turbine wake during various atmospheric conditions.