Fault recording data-based power system disturbance occurrence time identification method

Abstract

The invention relates to a fault recording data-based power system disturbance occurrence time identification method and belongs to the power system dispatching automation and power grid simulation technical field. The method includes the following steps that: a CUSUM algorithm is adopted to perform mutation data preliminary identification so as to obtain intervals in which signal mutation exists and alarm time points corresponding to each of the intervals; a waveform similarity test threshold value is set according to actual situation or experience, and similarity test is performed on waveform intervals in which continuous mutation exists, if a test result is greater than the threshold value, it can be determined that mutation of different degrees exists in two signal intervals, and identification is needed to be performed, otherwise, a latter mutation alarm signal is a false alarm signal and is deleted; and finally, a BG algorithm is adopted to perform fine identification on the intervals in which data mutation exists, and a mutation point with a most obvious mutation effect before an alarm time point is selected as a disturbance occurrence time point. The method of the invention is advantageous in simple implementation, fast calculation speed and suitability for field application.

Description

technical field [0001] The invention relates to a method for identifying the occurrence time of a power system disturbance based on fault wave recording data, and belongs to the technical fields of power system scheduling automation and power grid simulation. Background technique [0002] The construction of a smart grid requires the grid to be self-healing, and intelligent alarming of grid faults is an important prerequisite for realizing the "self-healing" of the grid. The local substation has a wealth of alarm data. Diagnosis and alarm in the substation can make full use of local information to obtain fast and reliable diagnosis results, which has inherent advantages. The current fault diagnosis methods mainly rely on the protection outlet information and switch action information, and the information used is very limited. In fact, relay protection engineers mainly rely on fault recording to restore the fault process and evaluate protection behavior on site. Therefore, ...

AI Technical Summary

Problems solved by technology

This type of method often faces the problem of "over-identification", that is, in order to reliably obtain all correct mutation points, the threshold is set lower, and then a large number of wrong mutation points are identified, which seriously interferes with the identification of the disturbance occurrence time.

Images