Distributed Fault Section Location for Active Distribution Network Based on Bayesian Complete Analytic Model

Abstract

With the expansion of the distribution network and the connection of distributed generations (DGs), the traditional centralized fault section location is difficult to meet the requirements of accuracy and reliability. To solve this problem, a distributed fault section location method for active distribution network (ADN) is proposed. The key points of this method are the Bayesian complete analytic model and the dimension reduction. Firstly, a distributed fault diagnosis model is constructed based on the multi-agent system (MAS), and the fundamental of fault section location is analyzed. Secondly, considering the missing and false positives of the feeder terminal unit (FTU), a Bayesian complete analytic model is proposed with the minimum variable dimension reduced by the fault contradiction theory. Finally, the whole process of the fault location method is directly given. Simulation results show that the method has high fault tolerance and can effectively solve the single point of failure (SPOF) problem in centralized algorithms, which is suitable for large-scale ADN.