A method for identifying and repairing abnormal data of an electronic current transformer

Abstract

The invention provides a method for identifying and repairing abnormal data of an electronic current transformer. When abnormal data are identified, an 'OR' relation is adopted to rewrite an originalwaveform-based continuous abnormal data criterion, meanwhile, the Kirchhoff current law is used as an auxiliary condition, and when the current phasor of a circuit connected to the same bus meets theKirchhoff current law, it is considered that no abnormal data exist; When the Kirchhoff current law is not met, it is determined that abnormal data exist, so that an abnormal data discrimination algorithm based on improved continuity is started to recognize and repair the abnormal data, and the abnormal data are repaired by using the two later items of data. According to the method, the situationof abnormal data missed judgment can be effectively avoided, meanwhile, the possibility that normal data adjacent to the abnormal data are misjudged to be abnormal data is effectively avoided, meanwhile, repair calculation is simple, and therefore the very high speed level can be achieved, and the requirement for the quick action performance of a relay protection system is met.

Description

technical field [0001] The invention relates to a method for identifying and repairing abnormal data of an electronic current transformer, belonging to the technical field of current transformer detection. Background technique [0002] Abnormal data of electronic current transformers may pose a huge threat to the normal operation of the relay protection system, which will have a great impact on the stable operation of the power grid, and will also bring economic losses and poor power consumption experience to users . Although the abnormal data of the electronic current transformer shows a sudden change in the sampling value and a peak in the waveform, its quality bit is marked as normal, which makes the secondary system of the substation still regard it as normal data This is also an important reason why abnormal data is difficult to identify and has great harm. The purpose of abnormal data identification is to accurately identify abnormal data, so as to temporarily block ...

AI Technical Summary

Problems solved by technology

Although the calculation accuracy of intelligent algorithms based on neural networks is very high, the calculation time is often close to one second (tested using MATLAB software on a computer with 4G memory 64-bit operating system and Intel i5 processor), This obviously does not meet the requirement of quick action

Images