Feature Reduction Techniques for Power System Security Assessment

Neural Networks (NN) have been applied to the security assessment of power systems and have shown great potential for predicting the security of large power systems. The curse of dimensionality states that the required size of the training set for accurate NN increases exponentially with the size of input dimension. Thus, an effective feature reduction technique is needed to reduce the dimensionality of the operating space and create a high correlation of input data with the decision space. This paper presents a new feature reduction technique for NN-based power system security assessment. The proposed feature reduction technique reduces the computational burden and the NN is rapidly trained to predict the security of power systems. The proposed feature reduction technique was implemented and tested on IEEE 50-generator, 145-bus system. Numerical results are presented to demonstrate the performance of the proposed feature reduction technique.

[1]  C. R. Rao,et al.  The Utilization of Multiple Measurements in Problems of Biological Classification , 1948 .

[2]  Craig A. Jensen,et al.  Application of computational intelligence to power system security assessment , 1999 .

[3]  Mohamed A. El-Sharkawi,et al.  Dynamic security contingency screening and ranking using neural networks , 1997, IEEE Trans. Neural Networks.

[4]  Robert J. Schalkoff,et al.  Pattern recognition - statistical, structural and neural approaches , 1991 .

[5]  R. Fisher THE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS , 1936 .

[6]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[7]  M.A. El-Sharkawi,et al.  Comparative study of feature extraction techniques for neural network classifier [power system simulation] , 1996, Proceedings of International Conference on Intelligent System Application to Power Systems.

[8]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[9]  Brian D. Ripley,et al.  Pattern Recognition and Neural Networks , 1996 .

[10]  C. W. Taylor,et al.  Recording and analyzing the July 2 cascading outage [Western USA power system] , 1997 .

[11]  M.A. El-Sharkawi,et al.  Use of Karhunen-Loe've expansion in training neural networks for static security assessment , 1991, Proceedings of the First International Forum on Applications of Neural Networks to Power Systems.

[12]  Vijay Vittal,et al.  Transient Stability Test Systems for Direct Stability Methods , 1992 .

[13]  C. Jensen,et al.  Power System Security Assessment Using Neural Networks: Feature Selection Using Fisher Discrimination , 2001, IEEE Power Engineering Review.