高级检索

便携式电流互感器多绕组极性自动校验装置的研制

Development of Automatic Polarity Verification Device for Multi-winding Portable Current Transformer

  • 摘要: 电流互感器多绕组极性校验方法主要通过向电流互感器一次侧通入直流电流,观察二次侧各绕组所接直流毫安表的指针偏转方向来判断极性,由于受到电磁干扰的影响,极性校验精度较差。对此,提出便携式电流互感器多绕组极性自动校验装置。采用二次侧电流检出及分析器采集电流互感器的二次侧电流信号,通过离散小波变换分解出不同的尺度系数,并结合逆变换操作实现信号重构,以降低电磁环境的干扰。利用Sigmoid函数拟合电流报警信号条件概率与母线电压有效值的关系,并通过最小二乘法确定函数参数。选取拟合函数的中心点反映报警敏感电压,以置信区间确定不确定域来量化不确定性。通过提取基频分量建立集合,并计算出比差值和角差值,以此校验极性。主机信号采集模块获取信号后,由PLC模块进行运算判断极性并控制辅机,显示与报警模块呈现结果。在实验中,对提出的方法进行了校验精度检验。最终测试结果表明,采用提出的方法进行电流互感器校验时,多绕组极性关联一致性系数均值为0.94,具备较为理想的校验效果。

     

    Abstract: The conventional method for polarity verification of current transformers with multiple windings primarily involves applying DC current to the primary side and observing the deflection direction of milliammeter pointers on secondary windings to determine polarity. However, this approach exhibits limited accuracy due to electromagnetic interference. To address this, a portable automatic polarity verification device for current transformers with multiple windings has been developed. The system employs a secondary current detector and analyzer to capture secondary current signals, decomposes them into different scale coefficients through discrete wavelet transform, and reconstructs signals via inverse transformation to mitigate electromagnetic interference. A Sigmoid function is utilized to model the relationship between current alarm signal conditions and effective bus voltage values, with parameters determined through least squares fitting. The central point of the fitted function reflects sensitive voltage thresholds, while confidence intervals quantify uncertainty ranges. By extracting fundamental frequency components and calculating ratio and phase differences, the system verifies polarity. The host's signal acquisition module processes data, with PLC modules performing computational verification of polarity, controlling auxiliary devices, and displaying results through display/alarm modules. Experimental validation demonstrated that the proposed method achieves an average correlation coefficient of 0.94 for multiple winding polarities, demonstrating optimal verification performance.

     

/

返回文章
返回