Analysis on Application of Wavelet Neural Network in Wind Electricity Power Prediction

Abstract:

Wind electricity power has fluctuation, and accurate and reasonable wind electricity power prediction is very important for solving wind electricity network and combination. This paper takes an analysis of a lot of actual data of a certain wind electricity field. Through wavelet neural network and time series method rolling, it can predict the overall power of wind electricity field. The result shows that for the original data of sampling time length and large sampling frequency, the model constructed by this paper has very good prediction effect. Because of the fan installation position, wind electricity fan flow effect and other random factor influence, wind electricity field overall power and single unit power distribution have difference. Through comparing with the time series parameters, it puts forward that single wind electricity unit power has smooth effect for overall power of wind electricity field. Finally, it summarizes the prediction effect and puts forward some reasonable suggestions for wind electricity network problems.

• Periodical: Applied Mechanics and Materials (Volume 686)
• Pages: 627-633
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.686.627
• Online since: October 2014
• Authors: Chun Yi Huang
• Keywords: Smooth Effect, Time Series, Wavelet Neural Network

References

[1] WANG Dexing. Analysis on AGC adjustment demand in electrical system operation[J]Electrical System Automation, 2004, 28(8).

[2] WANG Xiaochuan, SHI Feng, YU Lei, LI Yang. Analysis on Matlab 43 neural network cases[M]. Beijing Aerospace University Press, (2013).

[3] ZHANG Defeng et al. Matlab probability and mathematics statistics analysis[M]. Beijing: Mechanical Industrial Press, (2010).

[4] Kirby, B. &E. Hirst. Generator response to intrahour load fluctuations[J]. IEEE Transactions on Power Systems, 1998, 13(4): 1373-1378.

DOI: 10.1109/59.736279

[5] Peiyuan, C., T. Pedersen, B. Bak-Jensen. ARIMA-Based Time Series Model of Stochastic Wind Power Generation[J]. IEEE Transactions on Power Systems, 2010, 25(2): 667-676.

DOI: 10.1109/tpwrs.2009.2033277

[6] JIN Hao. Matlab application in timing sequence analysis [EB OL].