Ensemble neural network approach for accurate load forecasting in a power system

Krzysztof Siwek; Stanisław Osowski; Ryszard Szupiluk

International Journal of Applied Mathematics and Computer Science (2009)

  • Volume: 19, Issue: 2, page 303-315
  • ISSN: 1641-876X

Abstract

top
The paper presents an improved method for 1-24 hours load forecasting in the power system, integrating and combining different neural forecasting results by an ensemble system. We will integrate the results of partial predictions made by three solutions, out of which one relies on a multilayer perceptron and two others on self-organizing networks of the competitive type. As the expert system we will apply different integration methods: simple averaging, SVD based weighted averaging, principal component analysis and blind source separation. The results of numerical experiments, concerning forecasting the hourly load for the next 24 hours of the Polish power system, will be presented and discussed. We will compare the performance of different ensemble methods on the basis of the mean absolute percentage error, mean squared error and maximum percentage error. They show a significant improvement of the proposed ensemble method in comparison to the individual results of prediction. The comparison of our work with the results of other papers for the same data proves the superiority of our approach.

How to cite

top

Krzysztof Siwek, Stanisław Osowski, and Ryszard Szupiluk. "Ensemble neural network approach for accurate load forecasting in a power system." International Journal of Applied Mathematics and Computer Science 19.2 (2009): 303-315. <http://eudml.org/doc/207937>.

@article{KrzysztofSiwek2009,
abstract = {The paper presents an improved method for 1-24 hours load forecasting in the power system, integrating and combining different neural forecasting results by an ensemble system. We will integrate the results of partial predictions made by three solutions, out of which one relies on a multilayer perceptron and two others on self-organizing networks of the competitive type. As the expert system we will apply different integration methods: simple averaging, SVD based weighted averaging, principal component analysis and blind source separation. The results of numerical experiments, concerning forecasting the hourly load for the next 24 hours of the Polish power system, will be presented and discussed. We will compare the performance of different ensemble methods on the basis of the mean absolute percentage error, mean squared error and maximum percentage error. They show a significant improvement of the proposed ensemble method in comparison to the individual results of prediction. The comparison of our work with the results of other papers for the same data proves the superiority of our approach.},
author = {Krzysztof Siwek, Stanisław Osowski, Ryszard Szupiluk},
journal = {International Journal of Applied Mathematics and Computer Science},
keywords = {neural networks; blind source separation; ensemble of predictors; load forecasting},
language = {eng},
number = {2},
pages = {303-315},
title = {Ensemble neural network approach for accurate load forecasting in a power system},
url = {http://eudml.org/doc/207937},
volume = {19},
year = {2009},
}

TY - JOUR
AU - Krzysztof Siwek
AU - Stanisław Osowski
AU - Ryszard Szupiluk
TI - Ensemble neural network approach for accurate load forecasting in a power system
JO - International Journal of Applied Mathematics and Computer Science
PY - 2009
VL - 19
IS - 2
SP - 303
EP - 315
AB - The paper presents an improved method for 1-24 hours load forecasting in the power system, integrating and combining different neural forecasting results by an ensemble system. We will integrate the results of partial predictions made by three solutions, out of which one relies on a multilayer perceptron and two others on self-organizing networks of the competitive type. As the expert system we will apply different integration methods: simple averaging, SVD based weighted averaging, principal component analysis and blind source separation. The results of numerical experiments, concerning forecasting the hourly load for the next 24 hours of the Polish power system, will be presented and discussed. We will compare the performance of different ensemble methods on the basis of the mean absolute percentage error, mean squared error and maximum percentage error. They show a significant improvement of the proposed ensemble method in comparison to the individual results of prediction. The comparison of our work with the results of other papers for the same data proves the superiority of our approach.
LA - eng
KW - neural networks; blind source separation; ensemble of predictors; load forecasting
UR - http://eudml.org/doc/207937
ER -

References

top
  1. Afkhami-Rohani, K. R. and Maratukulam, D. (1998). ANNSTLF-Artificial neural network short-term load forecaster-Generation three, IEEE Transactions on Power Systems 13(4): 1413-1422. 
  2. Belouchrani, A., Abed-Meraim, K., Cardoso, J. and Moulines, E. (1997). A BSS technique using SOS, IEEE Transactions on Signal Processing 45(2): 434-444. 
  3. Choi, S., Cichocki, A. and Belouchrani, A. (2002). Second order nonstationary source separation, Journal of VLSI Signal Processing 32(1-2): 93-104. Zbl1010.94506
  4. Cichocki, A. and Amari, S. I. (2003). Adaptive Blind Signal and Image Processing, Wiley, New York, NY. 
  5. Cichocki, A., Amari, S., Siwek, K. and Tanaka, T. (2009). ICALAB Toolboxes, RIKEN, Tokyo, Available at: http://www.bsp.brain.riken.jp/ICALAB. 
  6. Cottrell, M., Girard, B., Girard, Y., Muller, C. and Rousset, P. (1995). Daily electrical power curve: Classification and forecasting using a Kohonen map, Proceedings of the International Workshop on Artificial Neutral Networks, IWANN, Malaga, Spain, pp. 1107-1113. 
  7. Diamantras, K. and Kung, S. Y. (1996). Principal Component Neural Networks, Wiley, New York, NY. 
  8. Drezga, I. and Rahman, S. (1998). Input variable selection for ANN-based short-term load forecasting, IEEE Transactions on Power Systems 13(4): 1238-1244. 
  9. Fidalgo, J. N. and Pecas Lopez, J. (2005). Load forecasting performance enhancement when facing anomalous events, IEEE Transations on Power Systems 20(2): 408-415. 
  10. Golub, G. and Van Loan, C. (1991). Matrix Computations, Academic Press, New York, NY. Zbl1118.65316
  11. Gonzalez-Romera, E., Jaramillo-Moran, M. A. and CarmonaFernandez, D. (2006). Monthly electric energy demand forecasting based on trend extraction, IEEE Transations on Power Systems 21(4): 1946-1953. 
  12. Guyon, I. and Elisseeff, A. (2003). An introduction to variable and feature selection, Journal of Machine Learning Research 3(3): 1158-1182. Zbl1102.68556
  13. Haykin, S. (2002). Neural Networks. A Comprehensive Foundation, Macmillan, New York, NY. Zbl0828.68103
  14. Hippert, H. S., Pedreira, C. E. and Souza, R. C. (2001). Neural networks for short-term load forecasting: A review and evaluation, IEEE Transations on Power Systems 16(1): 44-55. 
  15. Kandil, N., Wamkeue, R., Saad, M. and Georges, S. (2006). An efficient approach for short term load forecasting using artificial neural networks, Electrical Power and Energy Systems 28(4): 525-530. 
  16. Kiartzis, S. J., Zoumas, C. E., Theocharis, J., Bakirtzis, A. G. and Petridis, V. (1997). Short-term load forecasting in an autonomous power system using artificial neural networks, IEEE Transations on Power Systems 12(4): 1591-1596. 
  17. Kuntcheva, L. (2004). Combining Pattern Classifiers-Methods and Algorithms, Wiley, New York, NY. 
  18. Lendasse, A., Cottrell, M., Wertz, V. and Verleysen, M. (2002). Prediction of electric load using kohonen maps - Application to the Polish electricity consumption, Proceedings of the American Control Conference, Anchorage, AK, USA, Vol. 28, pp. 3684-3688. 
  19. Ljung, L. (1999). System Identification-Theory for the User, PTR Prentice Hall, Upper Saddle River, NJ. Zbl0615.93004
  20. Mandal, P., Senjyu, T., Urasaki, N. and Funabashi, T. (2006). A neural network based several hours ahead electric load forecasting using similar days approach, Electrical Power and Energy Systems 28(3): 367-373. 
  21. Nikias, L. and Petropulu, A. P. (1993). Higher-Order Spectral Analysis-A Nonlinear Signal Processing Framework, Prentice Hall, Upper Saddle River, NJ. Zbl0813.62083
  22. Osowski, S. (2006). Neural Networks for Information Processing, Warsaw University of Technology Press, Warsaw, (in Polish). 
  23. Osowski, S. and Siwek, K. (1999). The self-organizing neural network approach to load forecasting in power system, Proceedings of the International Joint Conference on Neural Networks, Washington, DC, USA, Vol. 28, pp. 1345-1348. 
  24. Osowski, S. and Siwek, K. (2002). Regularization of neural networks for load forecasting in power system, IEE Proceedings GTD 149(3): 340-345. 
  25. Sorjamaa, A., Hao, J., Reyhani, N., Li, Y. and Lendasse, A. (2007). Methodology for long-term prediction of time series, Neurocomputing 70(16-18): 2861-2869. 
  26. Yalcinoz, T. and Eminoglu, U. (2005). Short term and medium term power distribution load forecasting by neural networks, Energy Conversion and Management 46(8): 1393-1405. 

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

                
Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.