Ardema3D is a new type of vortex free model code that can simulate wind turbines. This code has been developed in order to replace the state of the art model, the blade element momentum (BEM). Indeed, the BEM is using strong assumptions and empirical correction, which are relevant for standard operation, however, there is a doubt on the validity of this model for complex situation. Ardema3D is developed to be substitute for the BEM when it is not valid. This code has been implemented in a coupled code, FARDEMAST, taking into account the elasticity of the blade and the controller. FARDEMAST is a code based on FAST, a code developed by the National Renewable Energy Laboratory (NREL), where the aerodynamic module has been replaced by Ardema3D. Both codes are under validation. The couple code FARDEMAST is computing the loads on the wind turbine, which are used to design the blade. Since vortex panel codes such as Ardema3D are using much less assumptions than the state of the art model for the industry, they are supposed to give more accurate results which can lead to safe and cheaper design for wind turbines. The Master Thesis proposed will cover two main topics of research : 1. Validation of the far wake model of Ardema3D The free wake vortex model ARDEMA3D is accurate in terms of description of the unsteady forces on the blades and on the rotor near wake velocities. The far wake description is not so accurate. For the wakes, Adwen and the CORIA have been developing an actuator line model inside the Large Eddy Simulation code YALES2 for very advanced and detailed wake simulations. Several validations of the actuator line model have been initiated with respect to the existing bibliography. The most useful experimental results considered the analysis of wake velocity deficits downstream of a small scale wind turbine placed on a wind tunnel in the Norwegian University of Science and Technology (NTNU) in Norway. Based on the NTNU experimental setup, the work to be performed during this Master Thesis will be to compare the results of Ardema3D and the actuator line model both in terms of local rotor quantities such as forces, angle of attacks, etc, on blades, where ARDEMA3D is assumed to be more accurate together with the velocity deficit close and far from the rotor where YALES2 should have a better description and analyze the sources of the differences. 2. Validation of the coupled code FARDEMAST After the validation of the aerodynamic code, the student will validate the use of the code for aero-elasticity purpose with the code FARDEMAST. The comparison will be done with FAST. The only difference with FARDEMAST is the use of an aerodynamic model based on the momentum equation, thus faster but less accurate. The validation will be done also with two structural models of FAST : ElastoDyn and BeamDyn. / Ardema3D ar en ny typ av vortexfri modellkod som kan simulera vindkraftverk. Denna kod har utvecklats for att ersatta den senaste tekniken, bladelementets momentum (BEM). Faktum ar att BEM anvander starka antaganden och empirisk korrigering, som ar relevanta for standardoperation, men det ar tveksamt om giltigheten av denna modell for komplexa situationer. Ardema3D ar utvecklad for att ersatta BEM nar den inte ar giltig. Denna kod har implementerats i en kopplad kod, FARDEMAST, med hansyn till bladets elasticitet och kontrollenheten. FARDEMAST ar en kod baserad pa FAST, en kod som utvecklats av National Renewable Energy Laboratory (NREL), dar den aerodynamiska modulen har ersatts av Ardema3D. Bada koderna ar under validering. Parkoden FARDEMAST beraknar lasterna pa vindturbinen, som anvands for att designa bladet. Eftersom vortexpanelkoder som Ardema3D anvander mycket mindre antaganden an toppmodernmodellen for industrin, ska de ge mer exakta resultat som kan leda till saker och billigare design for vindkraftverk. Den foreslagna uppsatsen ska omfatta tva huvudamnen av forskning: 1. Validering av Ardema3Ds langvaktmodell Fri vatsvortexmodellen ARDEMA3D ar korrekt nar det galler beskrivning av de ostadiga krafterna pa knivarna och pa rotorn nara vackthastigheter. Beskrivningen av langt kolvattnet ar inte sa exakt. For vackningarna har Adwen och CORIA utvecklat en manovermodell i Large Eddy Simulationkoden YALES2 for mycket avancerade och detaljerade vakningssimuleringar. Flera valideringar av manoverlinjemodellen har initierats med hansyn till den befintliga bibliografin. De mest anvandbara experimentella resultaten betraktade analysen av vackthastighetsbristerna nedstroms en liten vindturbin placerad pa en vindtunnel i Norges teknisk-naturvetenskapliga universitet (NTNU) i Norge. Utifran NTNU: s experimentella installning ar det arbete som ska utforas under denna examensarbete att jamfora resultaten fran Ardema3D och manovermodellen, bade nar det galler lokala rotorkvantiteter som krafter, angreppsvinkel mm pa blad, dar ARDEMA3D antas vara mer exakt tillsammans med hastighetsunderskottet nara och langt ifran rotorn dar YALES2 ska ha en battre beskrivning och analysera skillnadskallorna. 2. Validering av den kopplade koden FARDEMAST Efter validering av den aerodynamiska koden kommer studenten att validera anvandningen av koden for aeroelasticitetsandamal med koden FARDEMAST. Jamforelsen gors med FAST. Den enda skillnaden med FARDEMAST ar anvandningen av en aerodynamisk modell baserad pa momentumekvationen, sa snabbare men mindre noggrann. Valideringen kommer ocksa att goras med tva strukturella modeller av FAST: ElastoDyn och BeamDyn.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:kth-226319 |
| Date | January 2018 |
| Creators | Thibierge, Antoine |
| Publisher | KTH, Kraft- och värmeteknologi |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | TRITA-ITM-EX ; 2018:20 |
Page generated in 0.002 seconds