AI methods for fault location in MV networks

The fault location expert system is an application of a distribution management system (DMS), which is a part of the integrated control center environment composed of different computer systems and distribution automation. The expert system applies AI-based methods (e.g. object-oriented modelling, blackboard technique, fuzzy reasoning, decision tree) in incorporating data on external systems in order to support the control center operator in network operation. At present the expert system is in everyday use in several distribution utilities. The main contribution of the paper is to propose the expert system entity including advanced AI-based inference models. In addition to the inference models the novelty value is founded on the practical use of the system.