Wind turbines play an important role in providing electrical energy for an ever-growing demand. Due to climate change driven by anthropogenic emissions of greenhouse gases, the exploration and use of sustainable energy sources is essential with wind energy covering a significant portion. Data of existing wind turbines is needed to reduce the uncertainty of model predictions of future energy yields for planned wind farms. Due to maintenance routines and technical issues, data gaps of reference wind parks are unavoidable. Here, we present real-world case studies using multilayer perceptron networks and radial basis function networks to reproduce electrical energy outputs of wind turbines at 3 different locations in Germany covering a range of landscapes with varying topographic complexity. The results show that the energy output values of the turbines could be modeled with high correlations ranging from 0.90 to 0.99. In complex terrain, the RBF networks outperformed the MLP networks. In addition, rare extreme values were better captured by the RBF networks in most cases. By using wind meteorological variables and operating data recorded by the wind turbines in addition to the daily energy output values, the error could be further reduced to more than 20%. 1. Introduction The Combination of climate change and the dependence on fossil fuels slowly cause changes in energy policy and trigger an increasing demand for sustainable energy sources. Global carbon dioxide emissions are ever increasing and the associated consequences for the climate are widely scientifically recognized [1–3]. Over the last decade and in particular since the release of the report of the Intergovernmental Panel on Climate Change (IPCC) in 2007, public and political awareness of renewable energy technologies has increased considerably. This is not at least due to the large and fast growing economies and the associated increase of numbers of cars and energy consumption, and therefore of CO2 emissions [4]. Wind energy has the potential to be a vital contributor to renewable energy technologies that will substitute more and more for gas and coal [5]. In order to decrease the uncertainty of wind energy yield predictions during the planning of a single turbine or wind farm, data of nearby existing wind turbines are often used as reference for model evaluation. In Germany, the legislation that grants priority to renewable energy sources (Renewable Energy Resources Act, EEG) states that only wind turbines in areas with sufficient wind energy potential are qualified to receive compensation for
References
[1]
IPCC, “Climate Change 2013: The Physical Science Basis,” in Contribution of Working Group I To the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, T. F. Stocker, D. Qin, G. K. Plattner et al., Eds., p. 1535, Cambridge University Press, Cambridge, United Kingdom, 2013.
[2]
P. Friedlingstein, R. A. Houghton, G. Marland et al., “Update on CO2 emissions,” Nature Geoscience, vol. 3, no. 12, pp. 811–812, 2010.
[3]
G. P. Peters, G. Marland, C. le Quéré, T. Boden, J. G. Canadell, and M. R. Raupach, “Rapid growth in CO2 emissions after the 2008-2009 global financial crisis,” Nature Climate Change, vol. 2, no. 1, pp. 2–4, 2012.
[4]
G. P. Peters, C. L. Weber, D. Guan, and K. Hubacek, “China's growing CO2 emissions - A race between increasing consumption and efficiency gains,” Environmental Science and Technology, vol. 41, no. 17, pp. 5939–5944, 2007.
[5]
X. Lu, M. B. McElroy, and J. Kiviluoma, “Global potential for wind-generated electricity,” Proceedings of the National Academy of Sciences of the United States of America, vol. 106, no. 27, pp. 10933–10938, 2009.
[6]
EEG, “Renewable energy sources Act of 25 October 2008,” last amended by the Act of 20 December 2012, Art. 5G I, 2730, 2012.
[7]
I. Troen and E. L. Petersen, European Wind Atlas, Ris？ National Laboratory, Roskilde, Denmark, 1989.
[8]
N. G. Mortensen, L. Landberg, I. Troen, and E. L. Petersen, “Wind Atlas Analysis and Application Program (WAsP), Vol. 1: Getting Started. Vol. 2: User’s Guide,” Ris？ National Laboratory, Roskilde, Denmark, 1993.
[9]
R. Gasdch and J. Twele, Wind Power Plants: Fundamentals, Design, Construction and Operation, Springer, Berlin, Germany, 2nd edition, 2012.
[10]
W. Winkler, M. Strack, and A. Westerhellenweg, “Scaling and evaluation of wind data and wind farm energy yields,” DEWI Magazine, vol. 23, pp. 76–84, 2003.
[11]
FGW, “Technical Guidelines for Wind Turbines. Part 6. Determination of wind potential and energy yields,” Revision 8, Berlin, Germany, 2011.
[12]
C. M. Bishop, Neural Networks For Pattern Recognition, Oxford University Press, Oxford, UK, 1996.
[13]
D. W. Patterson, Artificial Neural Networks. Theory and Applications, Prentice Hall, Singapore, 1996.
[14]
D. S. Broomhead and D. Lowe, “Multi variable functional interpolation and adaptive networks,” Complex Systems, vol. 2, pp. 321–355, 1988.
[15]
B. D. Ripley, Pattern Recognition and Neural Networks, Cambridge University Press, Cambridge, UK, 1996.
[16]
S. Haykin, Neural Networks: A Comprehensive Foundation, Prentice Hall, Upper Saddle River, NJ, USA, 2nd edition, 1999.
[17]
S. Li, D. C. Wunsch, E. A. O'Hair, and M. G. Giesselmann, “Using neural networks to estimate wind turbine power generation,” IEEE Transactions on Energy Conversion, vol. 16, no. 3, pp. 276–282, 2001.
[18]
Z. Liu, W. Gao, Y. H. Wan, and E. Muljadi, “Wind power plant prediction by using neural networks,” in Proceedings of the IEEE Energy Conversion Conference and Exposition, Raleigh, NC, USA, September 2012.
[19]
W. A. Oost and E. M. Oost, “An alternative approach to the parameterization the momentum flux over the sea,” Boundary-Layer Meteorology, vol. 113, no. 3, pp. 411–426, 2004.
[20]
S. Ayoubi and K. L. Sahrawat, “Comparing multivariate regression and artificial neural network to predict barley production from soil characteristics in Northern Iran,” Archives of Agronomy and Soil Science, vol. 57, no. 5, pp. 549–565, 2011.
[21]
H. Z. Hamid Zare Abyaneh, “Evaluation of multivariate linear regression and artificial neural networks in prediction of water quality parameters,” Journal of Environ Health Science and Engineering, vol. 12, article 40, 2014.
[22]
D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning internal representations by error propagation,” in Parallel Distributed Processing: Explorations in the Microstructure of Cognition, D. E. Rumelhart, J. L. McClelland, and PDP Research Group, Eds., vol. 1, pp. 318–362, Cambridge University Press, Cambridge, Mass, USA, 1986.
[23]
E. Metcalfe, J. Teteh, and S. L. Howells, “Optimisation of radial basis and backpropagation neural networks for modelling auto-ignition temperature by quantitative-structure property relationships,” in Chemometrics and Intelligent Laboratory Systems, vol. 32, pp. 177–191, 1996.
[24]
M. Hestenes, Conjugate Direction Methods in Optimization, Springer, New York, NY, USA, 1980.
[25]
G. J. Bowden, H. R. Maier, and G. C. Dandy, “Optimal division of data for neural network models in water resources applications,” Water Resources Research, vol. 38, no. 2, pp. 2–11, 2002.
[26]
R. Khosla, “Knowledge-Based Intelligent Information and Engineering Systems,” in Proceedings of the KES 9th International Conference, Lecture Notes in Computer Science, p. 933, Melbourne, Australia, September 2005.
[27]
A. Schmidt, C. Hanson, J. Kathilankal, and B. E. Law, “Classification and assessment of turbulent fluxes above ecosystems in North-America with self-organizing feature map networks,” Agricultural and Forest Meteorology, vol. 151, no. 4, pp. 508–520, 2011.
[28]
B. Smith and H. Link, “Anemometer measurements for use in turbine power performance tests,” Preprint American Wind Energy Association (AWEA) Windpower Conference Portland, Oregon, USA, June 2002.
[29]
A. Albers and H. Sacman, “Wind speed and turbulence evaluation for performance and gondola anemometer measurements,” DEWI Magazine, vol. 10, pp. 51–62, 1997.
[30]
J. A. Anderson, An Introduction To Neural Networks, MIT Press, Boston, Mass, USA, 1995.
[31]
M. Qu, F. Y. Shih, J. Jing, and H. Wang, “Automatic solar flare detection using MLP, RBF, and SVM,” Solar Physics, vol. 217, no. 1, pp. 157–172, 2003.
[32]
A. Schmidt, T. Wrzesinsky, and O. Klemm, “Gap filling and quality assessment of CO2 and water vapour fluxes above an urban area with radial basis function neural networks,” Boundary-Layer Meteorology, vol. 126, no. 3, pp. 389–413, 2008.
[33]
Y. N. Maharani, S. Lee, and Y. K. Lee, “Topographical effects on wind speeds over various terrains: a case study for Korean peninsula,” in Proceedings of the 7th Asia-Pacific Conference on Wind Engineering, Taipei, Taiwan, November 2009.
[34]
M. P. Perrone and L. N. Cooper, “When networks disagree: ensemble methods for hybrid neural networks,” in Artificial Neural Networks For Speech and Vision, R. J. Mammone, Ed., pp. 126–147, Chapman & Hall/CRC, London, UK.
[35]
B. Bakker and T. Heskes, “Clustering ensembles of neural network models,” Neural Networks, vol. 16, no. 2, pp. 261–269, 2003.
[36]
K. J. Eidsvik, “A system for wind power estimation in mountainous terrain. prediction of askervein hill data,” Wind Energy, vol. 8, no. 2, pp. 237–249, 2005.
[37]
J. M. Zurada, A. Malinowski, and I. Cloete, “Sensitivity analysis for minimization of input data dimension for feedforward neural networks,” in Proceedings of the IEEE International Symposium on Circuits and Systems, pp. 447–450, IEEE Press, London, UK, February 1994.
[38]
R. E. Bellman, Adaptive Control Processes, Princeton University Press, Princeton, NJ, USA, 1961.
