A New Approach to Load Forecasting: Using Semi-parametric Method and Neural Networks

  • Abhisek Ukil
  • Jaco Jordaan
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4233)


A new approach to electrical load forecasting is investigated. The method is based on the semi-parametric spectral estimation method that is used to decompose a signal into a harmonic linear signal model and a non-linear part. A neural network is then used to predict the non-linear part. The final predicted signal is then found by adding the neural network predicted non-linear part and the linear part. The performance of the proposed method seems to be more robust than using only the raw load data.


Power System Load Forecast Complex Conjugate Pair Neural Network Toolbox Auto Regressive 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bitzer, B., Rösser, F.: Intelligent Load Forecasting for Electrical Power System on Crete. In: UPEC 1997 Universities Power Engineering Conference, UMIST-University of Manchester (1997)Google Scholar
  2. 2.
    Zivanovic, R.: Analysis of Recorded Transients on 765kV Lines with Shunt Reactors. In: Power Tech2005 Conference, St. Petersburg, Russia (2005)Google Scholar
  3. 3.
    Zivanovic, R., Schegner, P., Seifert, O., Pilz, G.: Identification of the Resonant-Grounded System Parameters by Evaluating Fault Measurement Records. IEEE Transactions on Power Delivery 19, 1085–1090 (2004)CrossRefGoogle Scholar
  4. 4.
    Jordaan, J., Zivanovic, R.: Time-varying Phasor Estimation in Power Systems by Using a Non-quadratic Criterium. Transactions of the South African Institute of Electrical Engineers (SAIEE) 95, 35–41 (2004), ERRATA 94(3), 171–172 (September 2004)Google Scholar
  5. 5.
    Gorry, P.: General Least-Squares Smoothing and Differentiation by the Convolution (Savitzky-Golay) Method. Analytical Chemistry 62, 570–573 (1990)CrossRefGoogle Scholar
  6. 6.
    Bialkowski, S.: Generalized Digital Smoothing Filters Made Easy by Matrix Calculations. Analytical Chemistry 61, 1308–1310 (1989)CrossRefGoogle Scholar
  7. 7.
    Draper, N., Smith, H.: Applied Regression Analysis, 2nd edn. John Wiley & Sons, Chichester (1981)MATHGoogle Scholar
  8. 8.
    Tibshirani, R.: Regression Shrinkage and Selection via the Lasso. Journal of the Royal Society. Series B (Methodological) 58, 267–288 (1996)MATHMathSciNetGoogle Scholar
  9. 9.
    Casar-Corredera, J., Alcásar-Fernándes, J., Hernándes-Gómez, L.: On 2-D Prony Methods, CH2118-8/85/0000-0796 $1.00, pp. 796–799. IEEE, Los Alamitos (1985)Google Scholar
  10. 10.
    Zivanovic, R., Schegner, P.: Pre-filtering Improves Prony Analysis of Disturbance Records. In: Eighth International Conference on Developments in Power System Protection, Amsterdam, The Netherlands (2004)Google Scholar
  11. 11.
    Wasserman, P.: Advanced Methods in Neural Computing. Van Nostrand Reinhold, New York (1993)MATHGoogle Scholar
  12. 12.
    Mathworks: MATLAB Documentation - Neural Network Toolbox. Version Release 13 edn. Mathworks Inc., Natick, MA (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Abhisek Ukil
    • 1
  • Jaco Jordaan
    • 1
  1. 1.Tshwane University of TechnologyPretoriaSouth Africa

Personalised recommendations