Gis-based Distribution Network Fault Location Study

With the development of electrical power system,the GIS has import academic and practial values to improve the reliablity of distribution network. The fault location is the important basic funciton of distribution GIS, it is meaningful to establish a effective fault location model. The paper analyzes the fault location problem on supervise area and non- supervise area, at last uses the methods based on Bayes formula and rough set theory.After deep study,this paper gets two research achivements as follows:First, considering the influence of the distorted information on fault location, the paper analyzes the fault location problem on supervise area,and uses a location method based on Bayes formula.This method handles SCADA fault information succinctly and shields distorted and error infromation. Fault probability and maximal ramarkable degree are adopted to establish fault location model.This location method is simple and effective.Second, after analyzing the fault location problem on non- supervise area,the paper suggests a method based on rough set theory. Rough set theory decision-making table is automatically created on the geometric network of distribution network, and a generic method of reducing decision-making table and searching minimum reduction is presented. It doesn't need any transcendental information for data set of the problem when rough set theory deal with uncertain problems, what's more it can reduce data then get minimum expression of knowledge under the precondition of assuring crucial information, ultimately the simple schema of information set is discovered, so the new fault location algorithm is very suitable for fault location on the non-supervise area.At last, the paper introduces the project background. Two achivements above are used in the GVMS. It proves the efficiency and feasibility of the mehods in both theroy and practice.