首页|基于局部放电超声信号的变压器绕组短路阻抗自动化测量技术

基于局部放电超声信号的变压器绕组短路阻抗自动化测量技术

扫码查看
为了预防变压器绕组变形,延长变压器的使用寿命,提出基于局部放电超声信号的变压器绕组短路阻抗自动化测量技术.获取变压器局部放电超声信号,采用小波提取变压器局部放电超声信号的特征;通过双向二维主成分分析算法(2DPCA)对冗余的特征信息进行降维,结合特征的能量变化率,检测变压器绕组短路状态;若变压器发生绕组短路,结合变压器等效电路模型自动化测量变压器绕组短路阻抗,采用降低互感器自动化测量误差的方法完成测量结果修正.实验结果表明,该方法自动化测量变压器绕组短路阻抗时的误差较小,负载率过大不会影响该方法的变压器绕组短路阻抗的自动化测量结果,同时可依据局部放电超声信号检测变压器绕组状态.
Automatic Measurement Technology of Transformer Winding Short Circuit Impedance Based on Partial Discharge Ultrasonic Signal
In order to prevent transformer winding deformation and extend the service life of the transformer,an auto-matic measurement technology for transformer winding short-circuit impedance based on partial discharge ultrasonic signals is proposed.Obtain the ultrasonic signal of transformer partial discharge,and use wavelet to extract the fea-tures of transformer partial discharge ultrasonic signal.By using the bidirectional two-dimensional principal component analysis algorithm(2DPCA)to reduce the dimensionality of redundant feature information and combining the energy change rate of features,the short-circuit status of transformer windings is detected.If a transformer experiences a winding short circuit,combined with the equivalent circuit model of the transformer,the short-circuit impedance of the transformer winding can be automatically measured,and the measurement results can be corrected by reducing the automatic measurement error of the transformer.The experimental results show that the error of this method in auto-matically measuring the short-circuit impedance of transformer windings is small,and an excessive load rate will not affect the automated measurement results of transformer winding short-circuit impedance using this method.At the same time,the transformer winding status can be detected based on partial discharge ultrasonic signals.

partial dischargetransformer windingshort-circuit impedancemutual inductorwavelet packet transform

许超、张书伟、薛瑞鹏

展开 >

张家口供电公司,张家口 075000

局部放电 变压器绕组 短路阻抗 互感器 小波包变换

2024

自动化与仪表
天津市工业自动化仪表研究所 天津市自动化学会

自动化与仪表

CSTPCD
影响因子:0.548
ISSN:1001-9944
年,卷(期):2024.39(3)
  • 10