Research Of Distribution Network's Fault Diagnosis Based On Data Mining

The aim of this thesis is to improve on a practical electric power distribution fault diagnosis system. A new electric power distribution fault diagnosis in accordance with data mining is proposed after dissertating the current power distribution system and its supporting system and analyzing application characteristic of various kinds of artificial intelligence in fault diagnosis.Firstly, the requirements and significance of electric power distribution fault diagnosis system and the current diagnosis methods as well as the disadvantages are discussed in the thesis. Data mining is taken as a major tool used in distribution network. The method of rough sets, artifical neural and decision tree in data mining are discussed.Rough Set theory is a powerful tool in dealing with vagueness and irrelevant information. It can be used to reduce features and extract rules. In this paper, rough set theory is applied to extract rules of distribution network. The test verifies that it is drastically effective.The distribution network's fault diagnosis is a complicated non-linear mapping. Neural networks are widely applied to pattern recognition. They can map complicated on-linear functions at infinite precision. In the paper, an improved BP networks are trained as a classifier of the distribution network's fault. Then, setting up a simple system of fault diagnosis about distribution by the Visual B and Matlab, and the results of an example are given.In the paper, decision tree method is firstly applied to realize the distribution network's fault diagnosis. Samples show that this method is effective for the distribution network's fault diagnosis.