The Application of Bayes Classifier in Power System Security Assessment

Abstract

This paper proposes a probabilistic method for power system security assessment, usingBayes classifier. The Bayes classifier, one of data classification tools, can apply to the power systemreliability area for calculations such as probabilistic security indices. The determination of securitybreach is a cumbersome and time-consuming process due to consideration of dynamic and steadystate effects. The straight Monte Carlo simulation, one of the commonly used methods in powersystem reliability, requires evaluation of each sampled state and can result in high computationtime. Once joint probability density of feature vectors is obtained, the Bayes classifier providesassessment of system security without complicated contingency analyses and can reduce thecomputational burden. Security status of a given feature vector can be determined by a Bayes rule,which can be implemented in power system reliability studies.