Recently, with growing application of wind power, the system based on the doubly fedinduction generator (DFIG) has become the one of the most popular concepts. Theproblem of connecting to the grid is also gradually revealed. As an effective solution toconnect offshore wind farm, VSC-HVDC line is the most suitable choice for stabilityreasons. However, there are possibilities that the converter of a VSC-HVDC link canadversely interact with the wind turbine and generate poorly damped sub-synchronousoscillations. Therefore, this master thesis will derive the linear model of a single DFIG aswell as the linear model of several DFIGs connecting to a VSC-HVDC link. For thelinearization method, the Jacobian transfer matrix modeling method will be explainedand adopted. The frequency response and time-domain response comparison betweenthe linear model and the identical system in PSCAD will be presented for validation. / Nyligen, med ökande tillämpning av vindkraft, det system som bygger på den dubbeltmatad induktion generator (DFIG) har blivit en av de mest populära begrepp. Problemetmed att ansluta till nätet är också gradvis avslöjas. Som en effektiv lösning för att anslutavindkraftpark är VSC -HVDC linje det lämpligaste valet av stabilitetsskäl. Det finns dockmöjligheter att omvandlaren en VSC-HVDC länk negativt kan interagera medvindturbinen och genererar dåligt dämpade under synkron svängningar. Därför kommerdetta examensarbete härleda den linjära modellen av en enda DFIG liksom den linjäramodellen av flera DFIGs ansluter till en VSC-HVDC -länk. För arise metoden kommerJacobian transfer matrix modelleringsmetodförklaras och antas. Jämförelse mellan denlinjära modellen och identiskt system i PSCAD frekvensgången och tidsdomänensvarkommer att presenteras för godkännande.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-12-22T11:06:45Z No. of bitstreams: 1 rodrigovianello.pdf: 847018 bytes, checksum: 7ad17011832e8156b0fa361091a1375a (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-12-22T12:53:13Z (GMT) No. of bitstreams: 1 rodrigovianello.pdf: 847018 bytes, checksum: 7ad17011832e8156b0fa361091a1375a (MD5) / Made available in DSpace on 2016-12-22T12:53:13Z (GMT). No. of bitstreams: 1 rodrigovianello.pdf: 847018 bytes, checksum: 7ad17011832e8156b0fa361091a1375a (MD5) Previous issue date: 2010-02-24 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O presente trabalho apresenta a proposição de duas metodologias inovadoras no campo da estimação fasorial. A primeira delas propõe a estimação do fasor da componente fundamental na presença de harmônicos e decaimento CC exponencial e é baseada na aplicação de filtros DFT (Discrete Fourier Transform) e de alguns métodos de processamento de sinais, tais como janelamento, modulação e DTFT (Discrete Time Fourier Transform). A segunda metodologia também propõe a estimação do fasor da componente fundamental, mas em um cenário mais complexo onde estão presentes no sinal, além dos harmônicos e decaimento CC exponencial, as oscilações subsíncronas. Esta metodologia é baseada na aplicação de redes neurais artificiais e de alguns métodos de processamento de sinais. Ambas as metodologias foram avaliadas frente a métodos tradicionais de estimação fasorial e apresentaram desempenhos superiores na presença de ruídos. / Thisthesispresents thepropositionoftwonovelmethodsin thefieldofphasor estimation. The first proposes to estimate the phasor of the fundamental component in the presence of harmonicsandexponentialdecayingDC.ThismethodologyisbasedontheapplicationoffiltersDFT (Discrete Fourier Transform) and some methods of signal processing such as windowing, modulation and DTFT (Discrete Time Fourier Transform). The second approach also proposes to estimate the phasor of the fundamental component, but in a more complex scenario formed for harmonics, exponential decaying DC and the sub-synchronous oscillations . This methodology isbasedontheapplicationofartificialneuralnetworksandsomemethodsofsignalprocessing. Both methods were evaluated against traditional methods of phasor estimation and the simulations showed that the two proposed methods were more accurate then others in presence of noise.