Application of Bayesian Learning Mechanism in Power System Transient Stability Assessment

Abstract:

One of the most challenging problems in real-time operation of power system is the prediction of transient stability. Fast and accurate techniques are imperative to achieve on-line transient stability assessment (TSA). This problem has been approached by various machine learning algorithms, however they find a class decision estimate rather than a probabilistic confidence of the class distribution. To counter the shortcoming of common machine learning methods, a novel machine learning technique, i.e. ‘relevance vector machine’ (RVM), for TSA is presented in this paper. RVM is based on a probabilistic Bayesian learning framework, and as a feature it can yield a decision function that depends on only a very fewer number of so-called relevance vectors. The proposed method is tested on New England power system, and compared with a state-of-the-art ‘support vector machine’ (SVM) classifier. The classification performance is evaluated using false discriminate rate (FDR). It is demonstrated that the RVM classifier can yield a decision function that is much sparser than the SVM classifier while providing higher classification accuracy. Consequently, the RVM classifier greatly reduces the computational complexity, making it more suitable for real-time implementation.

• Periodical: Advanced Materials Research (Volumes 108-111)
• Pages: 765-770
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.108-111.765
• Online since: May 2010
• Authors: Lin Niu, Jian Guo Zhao, Ke Jun Li, Zhen Yu Zhou
• Keywords: Bayesian Theorem, Probabilistic Learning, Relevance Vector Machine, Support Vector Machine (SVM), Transient Stability Assessment

References

[1] C. W. Taylor, The future in on-line security assessment and wide-area stability control, in Proc. 2000 IEEE Power Engineering Society Winter Meeting, pp.78-83.

DOI: 10.1109/pesw.2000.849931

[2] A. G. Phadke, Synchronized phasor measurements in power systems, IEEE Computer Applications in Power, 6(2), pp.10-15, Apr. (1993).

DOI: 10.1109/67.207465

[3] Ni Yixin, Theory and analysis of dynamic power systems, Tsinghua University Press, Beijing, (2002).

[4] L.S. Moulin, Support vector machines for transient stability analysis of large-scale power systems, IEEE Transactions on Power Systems, 19(2), pp.818-825, May. (2004).

DOI: 10.1109/tpwrs.2004.826018

[5] M.E. Tipping, Sparse Bayesian learning and the relevance vector machine, Mach Learning, 1, pp.211-244, (2001).

[6] N. Kaplowicz, Learning from imbalanced data sets: a comparison of various strategies, Proceeding of learning from Imbalanced Data Sets. AAAI Press. Menlo Park. CA. Tech. Rep. WS-00-05. 2000. PP. 10-15.

[7] Browne. M. W., Cross-validation methods, Journal of the Mathematical Psychology, vol. 44, pp.108-132, (2000).