Region-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain

We present a proof of concept of wind turbine wake identification and characterization using a region-based convolutional neural network (CNN) applied to lidar arc scan images taken at a wind farm in complex terrain. We show that the CNN successfully identifies and characterizes wakes in scans with...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jeanie A. Aird, Eliot W. Quon, Rebecca J. Barthelmie, Mithu Debnath, Paula Doubrawa, Sara C. Pryor
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	convolutional neural network wind turbine wakes lidar image processing wake characterization complex terrain Science Q
Acceso en línea:	https://doaj.org/article/36772c8fdc2b464cb6bde6b1ac612052
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:36772c8fdc2b464cb6bde6b1ac612052
record_format	dspace
spelling	oai:doaj.org-article:36772c8fdc2b464cb6bde6b1ac6120522021-11-11T18:56:44ZRegion-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain10.3390/rs132144382072-4292https://doaj.org/article/36772c8fdc2b464cb6bde6b1ac6120522021-11-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/21/4438https://doaj.org/toc/2072-4292We present a proof of concept of wind turbine wake identification and characterization using a region-based convolutional neural network (CNN) applied to lidar arc scan images taken at a wind farm in complex terrain. We show that the CNN successfully identifies and characterizes wakes in scans with varying resolutions and geometries, and can capture wake characteristics in spatially heterogeneous fields resulting from data quality control procedures and complex background flow fields. The geometry, spatial extent and locations of wakes and wake fragments exhibit close accord with results from visual inspection. The model exhibits a 95% success rate in identifying wakes when they are present in scans and characterizing their shape. To test model robustness to varying image quality, we reduced the scan density to half the original resolution through down-sampling range gates. This causes a reduction in skill, yet 92% of wakes are still successfully identified. When grouping scans by meteorological conditions and utilizing the CNN for wake characterization under full and half resolution, wake characteristics are consistent with a priori expectations for wake behavior in different inflow and stability conditions.Jeanie A. AirdEliot W. QuonRebecca J. BarthelmieMithu DebnathPaula DoubrawaSara C. PryorMDPI AGarticleconvolutional neural networkwind turbine wakeslidarimage processingwake characterizationcomplex terrainScienceQENRemote Sensing, Vol 13, Iss 4438, p 4438 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	convolutional neural network wind turbine wakes lidar image processing wake characterization complex terrain Science Q
spellingShingle	convolutional neural network wind turbine wakes lidar image processing wake characterization complex terrain Science Q Jeanie A. Aird Eliot W. Quon Rebecca J. Barthelmie Mithu Debnath Paula Doubrawa Sara C. Pryor Region-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain
description	We present a proof of concept of wind turbine wake identification and characterization using a region-based convolutional neural network (CNN) applied to lidar arc scan images taken at a wind farm in complex terrain. We show that the CNN successfully identifies and characterizes wakes in scans with varying resolutions and geometries, and can capture wake characteristics in spatially heterogeneous fields resulting from data quality control procedures and complex background flow fields. The geometry, spatial extent and locations of wakes and wake fragments exhibit close accord with results from visual inspection. The model exhibits a 95% success rate in identifying wakes when they are present in scans and characterizing their shape. To test model robustness to varying image quality, we reduced the scan density to half the original resolution through down-sampling range gates. This causes a reduction in skill, yet 92% of wakes are still successfully identified. When grouping scans by meteorological conditions and utilizing the CNN for wake characterization under full and half resolution, wake characteristics are consistent with a priori expectations for wake behavior in different inflow and stability conditions.
format	article
author	Jeanie A. Aird Eliot W. Quon Rebecca J. Barthelmie Mithu Debnath Paula Doubrawa Sara C. Pryor
author_facet	Jeanie A. Aird Eliot W. Quon Rebecca J. Barthelmie Mithu Debnath Paula Doubrawa Sara C. Pryor
author_sort	Jeanie A. Aird
title	Region-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain
title_short	Region-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain
title_full	Region-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain
title_fullStr	Region-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain
title_full_unstemmed	Region-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain
title_sort	region-based convolutional neural network for wind turbine wake characterization in complex terrain
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/36772c8fdc2b464cb6bde6b1ac612052
work_keys_str_mv	AT jeanieaaird regionbasedconvolutionalneuralnetworkforwindturbinewakecharacterizationincomplexterrain AT eliotwquon regionbasedconvolutionalneuralnetworkforwindturbinewakecharacterizationincomplexterrain AT rebeccajbarthelmie regionbasedconvolutionalneuralnetworkforwindturbinewakecharacterizationincomplexterrain AT mithudebnath regionbasedconvolutionalneuralnetworkforwindturbinewakecharacterizationincomplexterrain AT pauladoubrawa regionbasedconvolutionalneuralnetworkforwindturbinewakecharacterizationincomplexterrain AT saracpryor regionbasedconvolutionalneuralnetworkforwindturbinewakecharacterizationincomplexterrain
_version_	1718431636086849536

Region-Based Convolutional Neural Network for Wind Turbine Wake Characterization in Complex Terrain

Ejemplares similares