Fault characteristic extraction method for switch equipment based on big data platform

Abstract

The invention provides a fault characteristic extraction method for switch equipment based on a big data platform. According to the fault characteristic extraction method, the problem that the characteristic extraction cannot be performed efficiently and accurately on various types of faults when massive data of switch equipment faults in the prior art are faced is mainly solved. An implementation scheme of the method comprises the steps of establishing a hadoop sub-platform to carry out data collection, storage and pre-processing; establishing a SparkR platform to perform distributed calculation of MMSE (Multivariate Multi-Scale Entropy), and storing a calculation result into an HDFS (Hadoop Distributed File System); downloading the calculation result from the HDFS, and drawing a multivariate sample entropy curve of each fault of the switch equipment by use of R software; and selecting a multivariate sample entropy value in a corresponding scale factor range as a characteristic parameter of each fault according to the multivariate sample entropy curve of each fault. According to the fault characteristic extraction method for the switch equipment based on the big data platform, the whole scheme is designed precisely and completely, the capacities of massive data storage and distributed calculation are achieved, the efficiency and accuracy of the fault characteristic extraction are high, and a basis can be provided for diagnosis and prejudgment of the faults of the switch equipment in time.

Description

technical field [0001] The invention belongs to the technical field of industrial big data processing, and specifically relates to a switchgear fault feature extraction method, which can be applied to feature extraction of various faults of enterprise switchgear. Background technique [0002] As one of the terminal equipment of the power system, the switchgear shoulders the dual tasks of control and protection in the power system. Its reliability and intelligence level will have a profound impact on the stability and automation of the power system. Statistical analysis of switchgear accidents shows that the main causes of high-voltage circuit breaker failures are abnormal operating mechanism, SF6 leakage, damage to auxiliary parts and deterioration of main parts. The influencing factors of switchgear failure mainly include the service time of switchgear, annual load rate, environmental operation level, temperature, operation times and current times, etc. By extracting the f...

AI Technical Summary

Problems solved by technology

[0006] The technical idea of ​​the present invention is: through the processing of large sample data, the multi-variable and multi-scale entropy MMSE algorithm is introduced to solve the problem of increasing fault influencing factors, and indirectly improve the accuracy of fault feature extraction of switchgear; through the MMSE algorithm in Distributed parallel computing on the SparkR platform to improve the speed of feature extraction of switchgear faults

Images