Short-term Load Prediction of Power System Based on Genetic Neural Network

DOI

10.2991/iiicec-15.2015.110How to use a DOI?

Keywords

Genetic algorithm; Neural network; L-M algorithm;Power system; Load forecasting

Abstract

As the basic content of power system’s operation management and real-time control, the short-term load forecasting is significant for power system to run safely and economically. A short-term power system load predicting model based on genetic neural network is proposed, which take many factors into account, such as temperature, holidays and so on. In order to solve the problems of long convergence time and being easy to fall into local minimum of BP neural network, genetic algorithm (GA) is used to make a global searching for the initial weights and thresholds to solve the problems of long convergence time of BP neural network, and the Levenberg-Marquardt (L-M) method is used to train the network quickly. Based on the historical actual data, the forecasting results by the proposed method is more precise than those by BP neural network model, providing an effective way to forecast short-term power system load.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Shiwei Li
AU  - Yulong Li
PY  - 2015/03
DA  - 2015/03
TI  - Short-term Load Prediction of Power System Based on Genetic Neural Network
BT  - Proceedings of the 2015 International Industrial Informatics and Computer Engineering Conference
PB  - Atlantis Press
SP  - 478
EP  - 482
SN  - 2352-538X
UR  - https://doi.org/10.2991/iiicec-15.2015.110
DO  - 10.2991/iiicec-15.2015.110
ID  - Li2015/03
ER  -