• DocumentCode
    3210285
  • Title

    A method based on artificial neural network to estimate the health of wind turbine

  • Author

    Hui Li ; Jiarong Yang ; Menghang Zhang ; Shuangquan Guo ; Wei Lv ; Zongchang Liu

  • Author_Institution
    Central Academe, Shanghai Electr. Group Co., Ltd., Shanghai, China
  • fYear
    2015
  • fDate
    23-25 May 2015
  • Firstpage
    919
  • Lastpage
    922
  • Abstract
    This paper proposes a method based on the artificial neural network model to evaluate health state of wind turbine by using SCADA data. In this study, the core idea is to analysis the health condition of wind turbine by BP-ANN model, a kind of supervised learning technique is used in this proposed model, by selecting standard data as baseline data and compare with current testing data can realize the evaluation the state of wind turbines. By verifying the validation of the model through real SCADA data, and by visualization method and BP neural network to realize health assessment. This article focuses on the health status of the wind turbines, and provide a method to assess performance. Experimental results show that this method as an effective tool that can achieve the health assessment of wind turbines.
  • Keywords
    SCADA systems; backpropagation; condition monitoring; data visualisation; neural nets; power engineering computing; wind turbines; BP neural network; BP-ANN model; SCADA data; artificial neural network model; health assessment; health condition; performance assessment; supervised learning technique; visualization method; wind turbine health estimation; wind turbine health state evaluation; wind turbine health status; Artificial neural networks; Data models; Estimation; Prognostics and health management; Training; Wind turbines; ANN; Health Estimation; Wind Turbine;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Control and Decision Conference (CCDC), 2015 27th Chinese
  • Conference_Location
    Qingdao
  • Print_ISBN
    978-1-4799-7016-2
  • Type

    conf

  • DOI
    10.1109/CCDC.2015.7162050
  • Filename
    7162050