基于PCA-WPD优化的电流互感器故障检测方法研究

Research on fault detection method of current transformer based on PCA-WPD optimization

针对电磁式电流互感器测量误差的长期稳定性较差问题,提出主成分分析小波包分解的电流互感器故障测量误差自检测方法。利用小波包分解优化残差统计量,可以消除电流互感器设备中节点不平衡和随机误差对检测分析结果的影响。实验结果表明,随着测量误差的增加,测量数据的残差统计量逐渐增加。所提出的电流互感器故障检测方法能较好地满足0.2级精度的要求,且可以检测到电流互感器异常数据占90.97%,占总数的53.17%。该方法能够及时准确地实现电流互感器故障测量误差的自检测。

In order to solve the problem of poor long-term stability of electromagnetic current transformer measurement error,a principal component analysis wavelet packet decomposition method for fault measurement error self detection of current transformer was proposed.Using Wavelet packet decomposition to optimize residual statistics,the influence of node imbalance and random error in current transformer equipment on detection and analysis results could be achieved.The simulation experiment results showed that as the measurement error increased,the residual statistics of the measurement data gradually increased.The proposed current transformer fault detection method could better meet the requirement of 0.2 level accuracy,and could detect 90.97%of current transformer abnormal data,accounting for 53.17%of the total.This method could timely and accurately achieve self detection of current transformer fault measurement errors.