Advertisement

Data Imputation of Wind Turbine Using Generative Adversarial Nets with Deep Learning Models

  • Fuming Qu
  • Jinhai Liu
  • Xiaowei Hong
  • Yu Zhang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11301)

Abstract

Data missing problem is one of the most important issues in the field of wind turbine (WT). The missing data can lead to many problems that negatively affect the safety of power system and cause economic loss. However, under some complicated conditions, the WT data changes according to different environments, which would reduce the efficiency of some traditional data interpolation methods. In order to solve this problem and improve data interpolation accuracy, this paper proposed a WT data imputation method using generative adversarial nets (GAN) with deep learning models. First, conditional GAN is used as the framework to train the generative network. Then convolutional neural network is applied for both the generative model and the discriminative model. Through the zero-sum game between the two models, the imputation model can be well trained. Due to the deep learning models, the trained data imputation model can effectively recover the data with a few parameters of the input data. A case study based on real WT SCADA data was conducted to verify the proposed method. Two more data imputation methods were used to make the comparison. The experiments results showed that the method proposed in this paper is effective.

Keywords

Wind turbine Data interpolation Generative adversarial nets (GAN) Deep learning 

References

  1. 1.
    Craigmile, P.F., Guttorp, P.: Space-time modelling of trends in temperature series. J. Time Ser. Anal. 32(4), 378–395 (2011)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Oriani, F., et al.: Missing data simulation inside flow rate time-series using multiple-point statistics. Environ. Model. Softw. 86(C), 264–276 (2016)CrossRefGoogle Scholar
  3. 3.
    Shao, Y., Sun, Y., Liang, L.: Wind speed short-term forecast for wind farms based on ARMA model. Power Syst. Clean Energy 24(7), 52–55 (2008)Google Scholar
  4. 4.
    Zjavka, L.: Wind speed forecast correction models using polynomial neural networks. Renew. Energy 83, 998–1006 (2015)CrossRefGoogle Scholar
  5. 5.
    Petković, D., et al.: Estimation of fractal representation of wind speed fluctuation by artificial neural network with different training algorithms. Flow Meas. Instrum. 54, 172–176 (2017)CrossRefGoogle Scholar
  6. 6.
    Sun, W., Wang, Y.: Short-term wind speed forecasting based on fast ensemble empirical mode decomposition, phase space reconstruction, sample entropy and improved back-propagation neural network. Energy Convers. Manag. 157, 1–12 (2018)CrossRefGoogle Scholar
  7. 7.
    Goodfellow, I.J., et al.: Generative adversarial nets. In: International Conference on Neural Information Processing Systems, pp. 2672–2680. MIT Press (2014)Google Scholar
  8. 8.
    Chen, Y., et al.: Model-free renewable scenario generation using generative adversarial networks. IEEE Trans. Power Syst. PP(99), 1 (2017)Google Scholar
  9. 9.
    Li, J., et al.: WaterGAN: unsupervised generative network to enable real-time color correction of monocular underwater images, p. 99 (2017)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.College of Information Science and EngineeringNortheastern UniversityShenyangChina
  2. 2.Datang New Energy Experimental Research InstituteBeijingChina

Personalised recommendations