advanced
Journal Information
Journal Information

   Description
   Editorial Board
   Guide for Authors
   Ordering

Contents Services
Contents Services

   Regular Issues
   Special Issues
   Authors Index

Links
Links

   FEI STU Bratislava    deGruyter-Sciendo

   Feedback

[11-12, 2000] 

Journal of Electrical Engineering, Vol 51, 11-12 (2000) 296-300

SHORT-TERM ELECTRICAL DEMAND PROGNOSIS USING ARTIFICIAL NEURAL NETWORKS

Muhammad Riaz Khan - Čestmír Ondrůšek

   The paper discusses the implementation of artificial neural networks (ANN) approach for providing a structural framework for the representation, manipulation and utilization of historical load data and meteorological information concerning prediction of electrical power demand. The back-propagation technique based on the input/output mapping method has been chosen for structuring the neural network. The algorithm has been implemented and trained to predict the short-term power demand on an hourly basis. Test results from daily load forecasts for a large scale power system indicate that ANN can produce better results than traditional models such as time series and regression models.

Keywords: electrical load forecasting, temperature forecasting, artificial neural networks, back-propagation algorithm


[full-paper]


© 1997-2023  FEI STU Bratislava