Fault Diagnosis for Power System Transmission Line Based on PCA and SVMs - Intelligent Computing for Sustainable Energy and Environment - page 532

5Conluon
This paper demonstrated that the proposed PCA-SVM approach is a powerful
tool for monitoring power system transmission process. This methods is capable
of capturing the relationship between the recorded variables from the data, and
providing confidential limit charts for the violate fault points. It also helps to
extract the features of the faulty signal under different faulty situations, which
are used as the the inputs of the SVMs to classify these faults correctly.
The work however has also raised some important issues with respect to the
implementation of the combination method: (1) the monitoring statistics based
on linear PCA are not sucient to identify the nonlinear relations of the process
variables, therefore, it is necessary to develop nonlinear and dynamic extensions
for the nonlinear and dynamic system; (2) parameters of SVMs in this paper is
not optimized or selected to achieve precise classifications. So the work in the
near future is to solve these problems and to improve the PCA-SVM methods.
Acknowledgment. Thisworkwasfinanciallysupportedby RCUKincluding En-
gineering and Physical Sciences Research Council (EPSRC) under grant
EP/G042594/1,and partly supported by the China Scholarship Council, and Na-
tional Natural Science Foundation of China under Grants 61271347and 61273040.
Reference
[1] Salat, R., Osowski, S.: Accurate fault location in the power transmission line using
support vector machine approach. IEEE Transactions on Power Systems 19(2),
979-986 (2004)
[2] Frank, P.M.: Fault diagnosis in dynamic systems using analytical and knowledge-
based redundancy: A survey and some new results. Automatica 26(3), 459-474
(1990)
[3] Jolliffe, I.T., MyiLibrary: Principal component analysis, vol. 2. Wiley Online Li-
brary (2002)
[4] Venkatasubramanian, V., Rengaswamy, R., Yin, K., Kavuri, S.N.: A review of pro-
cess fault detection and diagnosis:: Part i: Quantitative model-based methods. Com-
puters & Chemical Engineering 27(3), 293-311 (2003)
[5] Dryden, I.L., Mardia, K.V.: Statistical shape analysis, vol. 4. John Wiley & Sons,
New York (1998)
[6] Kresta, J.V., Macgregor, J.F., Marlin, T.E.: Multivariate statistical monitoring
of process operating performance. The Canadian Journal of Chemical Engineer-
ing 69(1), 35-47 (1991)
[7] Suykens, J.A.K., Vandewalle, J.: Least squares support vector machine classifiers.
Neural Processing Letters 9(3), 293-300 (1999)
[8] Jakkula, V.: Tutorial on support vector machine (svm). School of EECS, Washing-
ton State University (2006)