Please use this identifier to cite or link to this item: https://doi.org/10.1109/ISAP.2007.4441660
Title: Electricity price forecasting using evolved neural networks
Authors: Srinivasan, D. 
Fen, C.Y.
Ah, C.L. 
Issue Date: 2007
Source: Srinivasan, D.,Fen, C.Y.,Ah, C.L. (2007). Electricity price forecasting using evolved neural networks. 2007 International Conference on Intelligent Systems Applications to Power Systems, ISAP : -. ScholarBank@NUS Repository. https://doi.org/10.1109/ISAP.2007.4441660
Abstract: Evolutionary techniques have capabilities of efficient search space exploration with population models corresponding to the problem. Their ability to capture the non linear dependencies among the system variables has invited economic analysts towards their use in the field of financial time series prediction. Although simple neural networks with sufficient number neuron units in the hidden layer are capable of following dynamics of any deterministic system, the weight search space becomes too complex to be searched using a simple back propagation based training algorithm. This paper presents and evaluates two alternative methods for finding the optimum weights of a neural network to capture the chaotic dynamics of electricity price data. The first method uses evolutionary algorithm to evolve a neural network, and the second method uses Particle Swarm Optimization for NN training. The global search capabilities of these evolutionary methods is used for finding the optimum neural network for forecasting electricity price from the California Power Exchange.
Source Title: 2007 International Conference on Intelligent Systems Applications to Power Systems, ISAP
URI: http://scholarbank.nus.edu.sg/handle/10635/70132
ISBN: 9860130868
DOI: 10.1109/ISAP.2007.4441660
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

6
checked on Dec 13, 2017

Page view(s)

19
checked on Dec 9, 2017

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.