Feature selection for dynamic stability prediction of power system using neural network
Corressponding author's email:
aunn@hcmute.edu.vnKeywords:
Dynamic stability prediction, power system, neural network, feature selectionAbstract
Dynamic stability prediction of power system faces with a large number of features, but not all features are useful. The redundant features will cause noise and reduce the perfor - mances of classifier. Feature selection aims to select a feature subset for classifier and improve recognition accuracy. This paper suggests the application of Relief algorithm for the feature selection. It is compared with two methods of feature selection that are Fisher discrimination and Divergence. Two recommended models for recognition accuracy are GRNN (Generalized Regression Neural Network) and MLPNN (Multilayer Perceptron Neural Network). Testing results on IEEE 39 - bus diagram show that the Relief algorithm with GRNN has yielded the results with smaller number of features and higher accuracy prediction than the other. Relief algorithm has significantly reduced the number of features while the recognition accuracy has been improved over all features.
Downloads: 0
References
M. R. Aghamohammadi, F. Mahdavizadeh, R. Bagheri, Power System Dynamic Security Classification Using Kohenen Neural Networks, 2009 IEEE, pp.1-7.
Rui Zhang, Yan Xu, Zhao Yang Dong and David J Hill, Feature Selection For Intelligent Stability Assessment of Power Systems, 2012 IEEE, pp.1-7.
C. A. Jensen and M. A. El-Sharkawi, Power system security assessment using neural net-works: feature selection using Fisher discrimination, IEEE Trans. Power Systems, vol. 16, Nov. 2001, pp.757-763.
K. Niazi, C. Arora, and S. Surana, Power system security evaluation using ann: feature selection using divergence, Electric Power Systems Research, vol. 69, Feb. 2004, pp.161- 167.
Taylor, Computational methods of feature selection, Edited by Huan Liu, Horoshi Mo- toda, 2008, pp.169-176.
I. S. Isa, Z. Saad, S. Omar, M. K. Osman, K. A. Ahmad, Suitable MLP Network Activa - tion Functions for Breast Cancer and Thyroid Disease Detection, 2010 IEEE, pp.39-44.
Matlab R2011b, Neural Network Toolboox User’s Guide.
Behnam Ababaei, Teymour Sohrabi and Farhad Mirzaei, Assessment of radial basis and generalized regression neural networks in daily reservoir inflow simulation, Elixir Comp. Sci. & Engg. 42, 2012, pp.6074-6077.
Simon S. Haykin, Neural Networks A Comprehensive Foundation, 2nd Edition, 1999, pp.84-86.
Quyen Huy Anh, The application of pattern recognition methods for fast analysis of the dynamic stability of electrical power system, Electrical technology, No. 2, pp.1-13, 1994, Perganon.
Hussain Hassan Al Marhoon, A Practical Method for Power Systems Transient Stability and Security, B.S. University of New Orleans, 2008, pp.56-60.
Prabha Kundur, Power System Stability and Control, New York: McGraw Hill, 1994, pp.17-39.
Downloads
Published
How to Cite
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Copyright © JTE.
