<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>The metering accuracy of the voltage transformer is related to the normal operation of the power system, and the metering results can be optimized through the prediction of the error state. In this paper, according to the generation mechanism of the measurement error of the transformer, the maximum information coefficient is used to extract the error characteristic quantity, and the measurement perturbation model is constructed by combining the ambient temperature and the secondary load factor. Due to the specificity of the ambient temperature, a BP neural network is also used to compensate for the temperature of the perturbation model, which prepares for the improved BP neural network based on a genetic algorithm to recognize the error data. Finally, the simulated operation of the three-phase voltage transformer and the measured data of the wiring substation were utilized for validation, respectively. With the help of three-phase CVT simulation, the error change of A-phase simulated CVT amplitude information at the 4001st sampling point is 0.0962%, and the error change of phase information is -4.572′.GA-BP neural network also has high sensitivity to the difficult-to-detect asymptotic error and is able to realize the error calibration of voltage transformer.</p>
</abstract>

The metering accuracy of the voltage transformer is related to the normal operation of the power system, and the metering results can be optimized through the prediction of the error state. In this paper, according to the generation mechanism of the measurement error of the transformer, the maximum information coefficient is used to extract the error characteristic quantity, and the measurement perturbation model is constructed by combining the ambient temperature and the secondary load factor. Due to the specificity of the ambient temperature, a BP neural network is also used to compensate for the temperature of the perturbation model, which prepares for the improved BP neural network based on a genetic algorithm to recognize the error data. Finally, the simulated operation of the three-phase voltage transformer and the measured data of the wiring substation were utilized for validation, respectively. With the help of three-phase CVT simulation, the error change of A-phase simulated CVT amplitude information at the 4001st sampling point is 0.0962%, and the error change of phase information is -4.572′.GA-BP neural network also has high sensitivity to the difficult-to-detect asymptotic error and is able to realize the error calibration of voltage transformer.

Voltage transformer metering error state prediction method based on GA-BP algorithm

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

The metering accuracy of the voltage transformer is related to the normal operation of the power system, and the metering results can be optimized through the...

Voltage transformer metering error state prediction method based on GA-BP algorithm

Pubblicato online: 06 dic 2023

Pagine: -

Ricevuto: 16 gen 2023

Accettato: 16 giu 2023

DOI: https://doi.org/10.2478/amns.2023.2.01385

Parole chiave
Voltage transformer, Maximum information coefficient, Disturbance model, BP neural network, Genetic algorithm, Temperature compensation

© 2023 Shuai Gao et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Voltage transformer metering error state prediction method based on GA-BP algorithm

Pubblicato online: 06 dic 2023

Pagine: -

Ricevuto: 16 gen 2023

Accettato: 16 giu 2023

DOI: https://doi.org/10.2478/amns.2023.2.01385

Parole chiaveVoltage transformer, Maximum information coefficient, Disturbance model, BP neural network, Genetic algorithm, Temperature compensation

© 2023 Shuai Gao et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Parole chiave
Voltage transformer, Maximum information coefficient, Disturbance model, BP neural network, Genetic algorithm, Temperature compensation