The Analysis And Application Of Association Rule And Decision Tree Algorithm In The Grid Fault

With the development of the national economy,people's production and living with electrical specifications and security levels level up, the regional power grid has become closer. The increasing of abnormal information of the power and the frequent occurrence of power system accidents are the factors that restrict the development of the country. The accidents of power system in severe weather show diversity and complexity, and the connections between the various properties seemingly unrelated, many potential strong association rules and decision patterns can be found by means of data mining. Cases of real power system operation of fault information mining, for power system fault prediction and early-warning system provides scientific basis, for the stable running of power system continuously is very meaningful.In this paper, through the detailed analysis of theory and data source of a provincial power grid 110 kV and above overhead transmission line accident, and a historical analysis of the operational characteristics of the power network under the influence of bad weather conditions is carried out. Constructing a model to mining under different weather grid disaster type and the impact extent and mining by forming the optimized multidimensional association rules algorithm and decision tree algorithm. Using the algorithm can extract the valuable information from the numerous and complicated historical grid fault information,to prevent and reduce the power grid accident occurrence and rapid classification of power network fault information, play a predictive role. The method used to solve the grid fault information mining and transaction data set in the power system changing problems, hoping to the provincial power grid power grid construction and efficient development in the future provided valuable experience and theoretical support. specific projects as follows.Firstly, this paper introduced a model of fast prediction in the gird fault. the provincial power grid in the past five years in the actual operating process of the line resulting in the factors of the failure of the grid lines were analyzed and summarized, and in accordance with the power system index system to construct transaction data model.Secondly,this paper proposed a new mode of combination association rules and the decision tree algorithm.The new property is formed by the optimized association rulealgorithm, and the data set is reorganized..Partition the data set into decision tree algorithm..Make up the decision tree in the process of constructing over fitting defects, solve problem of changing in different periods of the same transaction power system fault data source.Thirdly, through the example analysis, the algorithm is applied to the actual power grid fault diagnosis to environmental analysis, classifying data, model application and analysis results to verify the feasibility and effectiveness.