Current Transformer Saturation Waveform Recovery Method Based on Model and Data Hybrid Drive

Abstract

The invention discloses a current transformer saturation waveform recovery method based on model and data hybrid drive, comprising the following steps: S1, building a database; S2, building and training a fault current model key parameter identification network based on a long short-term memory network; S3 . Calculate the real fault current based on the identification network in step S2; S4, take the real fault current obtained in step S3 as the action information of the protection element, and wait for the next fault to occur. Taking the physical model of fault current as prior knowledge, a data-driven method is used to build a network for identifying key parameters of fault current model to achieve waveform recovery, with low sampling rate requirements, strong anti-noise capability, no need to set thresholds, and easy online / offline deployment and other advantages.

Description

technical field [0001] The invention relates to the technical field of saturation identification and waveform recovery of current transformers, and more particularly to a method for recovering saturation waveforms of current transformers based on hybrid driving of models and data. Background technique [0002] At present, the current transformer is an important measuring element in the power system, and its working performance directly affects the reliability of the action of the relay protection element. Due to the magnetic saturation characteristics of the iron core, the P-class current transformers widely used in 220kV and below systems have poor tolerance to the DC component in the fault current, and the saturation problem is unavoidable. When the current transformer is saturated, its transmission characteristics change from linear to nonlinear, so that the measured value cannot accurately reflect the real fault current, which may lead to malfunction and rejection of pro...

AI Technical Summary

Problems solved by technology

[0003] However, the method based on mutation information is susceptible to noise interference, and generally requires a higher sampling rate

In order to improve the anti-noise ability, the threshold value of mutation information detection can be increased appropriately, but this will lead to the problem of insufficient sensitivity when the saturation degree of current transformer is shallow

[0004] Therefore, there is a contradiction between the sensitivity and reliability of the method based on mutation information, making it difficult to select an appropriate threshold is an urgent problem to be solved by those skilled in the art

Images