Fast Monte Carlo reliability evaluation using support vector machine

Abstract This paper deals with the feasibility of using support vector machine (SVM) to build empirical models for use in reliability evaluation. The approach takes advantage of the speed of SVM in the numerous model calculations typically required to perform a Monte Carlo reliability evaluation. The main idea is to develop an estimation algorithm, by training a model on a restricted data set, and replace system performance evaluation by a simpler calculation, which provides reasonably accurate model outputs. The proposed approach is illustrated by several examples. Excellent system reliability results are obtained by training a SVM with a small amount of information.

[1]  E. Reingold,et al.  Combinatorial Algorithms: Theory and Practice , 1977 .

[2]  Wenyuan Li,et al.  Reliability Assessment of Electric Power Systems Using Monte Carlo Methods , 1994 .

[3]  G. Cojazzi,et al.  Using Neutral Networks as nonlinear model interpolators in variance decomposition-based sensitivity analysis , 2001 .

[4]  Roy Billinton,et al.  Reliability evaluation of engineering systems : concepts and techniques , 1992 .

[5]  Bartolomeo Sapio,et al.  SEARCH (Scenario evaluation and analysis through repeated cross impact handling): a new method for scenario analysis with an application to the Videotel service in Italy , 1995 .

[6]  Nello Cristianini,et al.  Controlling the Sensitivity of Support Vector Machines , 1999 .

[7]  I C G Campbell,et al.  Radial Basis Function Networks: Design and Applications , 2000 .

[8]  W. J. Burke,et al.  Creating data bases for power systems planning using high order linear interpolation , 1988 .

[9]  Colin Campbell,et al.  An introduction to kernel methods , 2001 .

[10]  Bernhard Schölkopf,et al.  Statistical Learning and Kernel Methods , 2001, Data Fusion and Perception.

[11]  Nello Cristianini,et al.  An introduction to Support Vector Machines , 2000 .

[12]  M.V.F. Pereira,et al.  A new computational tool for composite reliability evaluation , 1992 .

[13]  Y. B. Yoo,et al.  A comparison of algorithms for terminal-pair reliability , 1988 .

[14]  G. Shrestha,et al.  Analysis of inconsistent data in power planning , 1991, IEEE Power Engineering Review.

[15]  Colin Campbell,et al.  Kernel methods: a survey of current techniques , 2002, Neurocomputing.