Grid fault diagnosis method based on data mining

Abstract

The invention discloses a grid fault diagnosis method based on data mining. On the basis of the characteristics that grid measurement data quantity is large and dimensionality is high, grid fault diagnosis is achieved by means of a data mining technique. Firstly, the data correlation degree is evaluated by means of a neighborhood rough set method, reduction processing is carried out on data by means of a greedy search algorithm, then failure diagnoses are classified according to features by means of a competition coagulation clustering algorithm, and then failure diagnosis of a grid is achieved on the basis of a fuzzy association rule algorithm. According to the method, failure detection, classification and positioning of various devices and elements which break down in the grid with mass data can be carried out, ultimately, an IDEA integrated development tool and an MySQL database are used, an object-oriented program design concept is adopted, and a grid failure diagnosis system in a B / S mode is developed.

Description

technical field [0001] The invention relates to the field of power grid scheduling and fault analysis methods, in particular to a data mining-based power grid fault diagnosis method. Background technique [0002] Power grid fault diagnosis is to use the protection and switch action information in SCADA, combined with the principle of relay protection to identify faulty components and protection and switches that refuse to operate or malfunction. With the improvement of the voltage level of the power grid and the access of distributed power sources, the fault information of the distribution network is becoming more and more complicated; Uncertain faults in power grids such as power grids increase the difficulty of fault diagnosis in distribution networks, which leads to the failure of traditional diagnostic methods based on relay protection action information to achieve satisfactory dispatching results. [0003] Existing fault diagnosis methods mainly include expert system m...

AI Technical Summary

Problems solved by technology

In recent years, some scholars have considered from the three aspects of time dimension, space dimension and information dimension, and proposed an online fault diagnosis method based on multi-dimensional data, but it can only roughly analyze the fault differences of faulty equipment; However, the existing grid fault diagnosis method based on the causal rule network only uses the protection and switching action information of the station, which is easily affected by unilateral protection misoperation and misjudges the normal line as a fault; Based on the rule network, a set operation-based judgment protection and switch misoperation and refusal method are proposed, which improves the accuracy of fault diagnosis, but ignores the real-time requirements of massive data processing

[0004] Therefore, in view of the large amount of power grid measurement data, high dimensionality, and complex internal and external interference, an accurate, real-time, and robust power grid fault diagnosis method is urgently needed to make up for the shortcomings of existing technologies.

Images