A New Uncertain Fault Diagnosis Approach of Power System Based on Markov Chain Monte Carlo Method

Abstract

In this paper, a new fault diagnosis approach in large scale power grid based on Bayesian network and MCMC method is proposed for large scale power grid. Tow models of Bayesian network for constructing the Bayesian network of power grid are established. The main idea for Bayesian network approach is to compute the posterior probabilities of the fault nodes of the Bayesian network in MCMC method so that the fault in the power grid can be diagnosed. With the capacity of revealing relationships among data in model mentioned above, this approach highly improves the accuracy of fault diagnosis and is especially suitable for those environments with imperfect and uncertain information. Results of the testing example prove that the approach proposed is correct, effective and has potential for application of real-time fault diagnosis.