A probabilistic fault localization method for distribution networks using fault indicators

This paper presents a new method for fault localization on distribution networks. The developed method aims at improving the fault localization process by remotely retrieving fault indicators´ information and analyzing it with a probabilistic algorithm robust to corrupted data. On fault occurrence the developed algorithm proposes to the operator the most probably faulty section and detects the devices which may be defective, hence greatly enhancing the localization process.