Return to search

Optimering av en flytande vindkraftspark

This paper presents a study on the optimization of floating offshore wind farms. The aim of this study is mainly to create a tool that can help determine the most profitable layout option for the floating offshore wind power company Windeed. This report contains an overview of wind power theory including wind roses, wake losses, fatigue loading, and construction theory. The methodology used for modeling and optimization is the programming language Python together with additional tools such as TopFarm and PyWake. Challenges in the process of designing the layout of a floating wind farm are discussed and the two mathematical models, Bastankhah Gaussian and NOJ, are compared for their ability to recreate wake effects. In the results, the discoveries of the study and the tool created for the company are presented. It was found that allowing freely placed turbines rather than placing the turbines in a strict hexagonal pattern, with shared anchors, gave notably higher annual power production for examined wind farms. Although the levelized cost of energy of the farms with hexagonal patterns were lower for some of the investigated scenarios. Some of the key factors that need to be considered when choosing layout for an off shore wind farm, as well as the potential improvements of the tool are also highlighted in the discussion chapter. Overall, this study provides valuable insights into the design and optimization of floating offshore wind farms.

Identiferoai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-503778
Date January 2023
CreatorsBlavier, Mattias, Granath, Elin, Jin, Emelie, Johansson, Elin, Svensson, Axel, Thylin, Kristina
PublisherUppsala universitet, Materialteori
Source SetsDiVA Archive at Upsalla University
LanguageSwedish
Detected LanguageEnglish
TypeStudent thesis, info:eu-repo/semantics/bachelorThesis, text
Formatapplication/pdf, application/pdf
Rightsinfo:eu-repo/semantics/openAccess, info:eu-repo/semantics/openAccess
RelationFYSAST ; FYSPROJ1311

Page generated in 0.0023 seconds