Por favor, use este identificador para citar o enlazar este ítem: https://hdl.handle.net/10495/25979
Registro completo de metadatos
Campo DC Valor Lengua/Idioma
dc.contributor.advisorJaramillo Duque, Álvaro-
dc.contributor.advisorCano Quintero, Juan Bernardo-
dc.contributor.authorPérez González, Andrés Fernando-
dc.date.accessioned2022-02-11T19:45:15Z-
dc.date.available2022-02-11T19:45:15Z-
dc.date.issued2022-
dc.identifier.urihttp://hdl.handle.net/10495/25979-
dc.description.abstractABSTRACT : This report proposes a strategy that determines the coverage path in unmanned air vehicles over Photovoltaic (PV) plants. The report evaluates three image analysis methods for delineating PV plants from images in a database, also assesses three coverage path planning methods, and checks the coverage path planning methods implemented in two UAVs to cover three PV plants in the Gazebo simulation environment. The manuscript consists of 8 chapters and two of these are previously published journal articles. The first chapter evaluates some of the image analysis or computer vision methods for the delimitation of photovoltaic plants from a database, existing in the literature of the state of the art. The use of a convolutional neural network structure is proposed to perform the extraction of the PV plant (semantic segmentation) using the u-net model and then comparing the results with the models proposed by other authors. Finding that the u-net model has a more robust performance according to the most used standard metrics for the task, with respect to the other proposed methods. The second chapter specified a method for coverage path planning in geometric areas of photovoltaic plants. Three CPP methods with several performance metrics were programmed, three simulate PV plants were modeled and two UAVs were selected. Six experiments were carried out with each PV plant varying the CPP width. From the results obtained in these experiments, one of the metrics was selected and interpolated with respect to energy consumption, a study of energy consumption was carried out for each of these possibilities, where the CPP with the lowest energy consumption is the Boustrophedon exact cell decomposition when it has the widths in a range between 0 and 7 meters, on the other hand, the Grating Based Wavefront Coverage methods has the lowest energy consumption when the CPP width is greaterthan 7 meters. Concluding that UAVs with a short flight time from 10 to 15 minutes, can be used for the inspection of small PV plants (such as roofs, rooftops, canopies, and facades), where energy consumption is enough to do the inspection and return home, as well as, UAVs such as the Typhoon with up to 25 minutes of flight are appropriate for larger photovoltaic plants, also some other results were obtained from the experiments in Gazebo simulation environment. The information generated with this work could be valuable for the inspection of photovoltaic plants, with cameras of UAV, whose resolution is less than 12 Megapixels, in addition, the results obtained served to identify a set of recommendations for future work with UAVs in Operation and maintenance (O&M) of PV plants.spa
dc.format.extent74spa
dc.format.mimetypeapplication/pdfspa
dc.language.isoengspa
dc.type.hasversioninfo:eu-repo/semantics/draftspa
dc.rightsinfo:eu-repo/semantics/openAccessspa
dc.rights.urihttp://creativecommons.org/licenses/by-nc-sa/2.5/co/*
dc.titleCoverage Path Planning with Unmanned Aerial Vehicles over Photovoltaic Plants using Image Analysisspa
dc.title.alternativePlanificación de trayectorias de cobertura con vehículos aéreos no tripulados sobre plantas fotovoltaicas mediante el análisis de imágenesspa
dc.typeinfo:eu-repo/semantics/masterThesisspa
dc.publisher.groupGrupo de Manejo Eficiente de la Energía (GIMEL)spa
dc.description.noteTESIS CON DISTINCIÓN: Summa Cum Laude (Excelente)spa
oaire.versionhttp://purl.org/coar/version/c_b1a7d7d4d402bccespa
dc.rights.accessrightshttp://purl.org/coar/access_right/c_abf2spa
thesis.degree.nameMagíster en Ingenieríaspa
thesis.degree.levelMaestríaspa
thesis.degree.disciplineFacultad de Ingeniería. Maestría en Ingenieríaspa
thesis.degree.grantorUniversidad de Antioquiaspa
dc.rights.creativecommonshttps://creativecommons.org/licenses/by-nc-sa/4.0/spa
dc.publisher.placeMedellínspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdccspa
dc.type.redcolhttps://purl.org/redcol/resource_type/TMspa
dc.type.localTesis/Trabajo de grado - Monografía - Maestríaspa
dc.subject.lembAprendizaje automático (inteligencia artificial)-
dc.subject.lembMachine learning-
dc.subject.agrovocUnmanned aerial vehicles-
dc.subject.agrovocVehículos aéreos no tripulados-
dc.subject.agrovocImage processing-
dc.subject.proposalDeep Learningspa
dc.subject.proposalImage segmentationspa
dc.subject.proposalsemantic segmentationspa
dc.subject.proposalPhotovoltaic plantspa
dc.subject.proposalCoverage Path Planningspa
dc.subject.agrovocurihttp://aims.fao.org/aos/agrovoc/c_3eb20052-
dc.subject.agrovocurihttp://aims.fao.org/aos/agrovoc/c_37359-
dc.relatedidentifier.urlhttps://www.mdpi.com/2076-3417/11/14/6524spa
dc.relatedidentifier.urlhttps://www.mdpi.com/2076-3417/11/24/12093spa
Aparece en las colecciones: Maestrías de la Facultad de Ingeniería

Ficheros en este ítem:
Fichero Descripción Tamaño Formato  
PérezAndres_2022_CPPwithUAVoverPVplants.pdfTesis de maestría13.33 MBAdobe PDFVisualizar/Abrir


Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons Creative Commons