Posters

Abstract

The effective use of vertical profiling lidar in wind energy applications has become standard. However, there remains considerable potential for improving turbulence measurements [1]. Further optimization opportunities exist regarding data evaluation routines and processing of wind measurement data. This contribution presents initial results of a turbulence correction for vertical profiling lidar using machine learning methodologies. Accurate turbulence measurement at wind farm sites, particularly in complex geographical contexts, remains a significant challenge for vertical profiling lidar devices. We aim to qualify these devices for enhanced accuracy in turbulence assessments, thereby solidifying their position in wind energy research and market applications. The development of an AI-based module for turbulence correction represents an innovative and substantial advancement in addressing these challenges. The influence of vectorial and scalar averaging on turbulence is analyzed at various measurement locations. In addition, a correction method for the turbulence is developed using AI. Different modelling techniques were examined and tested for this purpose, such as Linear Regression, Random Forest, Variance Ratio and Support Vector Machine.  Lidar and measurement mast data were used for developing the turbulence module. This data was collected as part of the project from specially designed measurement campaigns in complex terrain. This provided an optimal basis for testing and developing the module. A user-friendly modular framework will facilitate these objectives, and the methods developed will be benchmarked against established techniques to ascertain their effectiveness.  The methods developed will ultimately be integrated into a modular framework, contributing to the research project “LoTar - Lidar data correction for sites in complex terrain” funded by the German Federal Ministry for Economic Affairs and Climate Action (BMWK). The project focuses on creating a comprehensive data evaluation system for lidar measurements. Within this framework, modules aimed at enhancing the processing of vertical profiling lidar data in complex terrains and improving turbulence measurements are under development. The framework is intended to be published after the LoTar project ends and to be widely applied. To ensure this, the work is carried out in close collaboration with stakeholders from the wind resource assessment sector. [1] A. Sathe, J. Mann, J. Gottschall, and M. S. Courtney, ‘Can Wind Lidars Measure Turbulence?’, Journal of Atmospheric and Oceanic Technology, vol. 28, no. 7, pp. 853–868, Jul. 2011, doi: 10.1175/JTECH-D-10-05004.1