提出了一种基于参数识别的电流互感器(CT)饱和识别原理,通过对系统区外故障时的故障分量网络进行分析得知,保护安装处的电压和电流均满足一个等效的电感电路模型。当电流互感器没有饱和时,利用最小二乘法识别出的模型参数反映系统参数的真实值且数值稳恒不随时间变化;一旦电流互感器发生饱和,利用该侧电流识别出的模型参数的数值会出现剧烈波动,通过计算反映参数波动程度大小的离散度即可准确判别出CT饱和。新原理本身对于非周期分量以及谐波都适用,不受系统运行方式以及故障初相角的影响,在应用时判据易整定;基于时域的算法,判别速度快,具有选择性,能够准确定位出饱和CT,而且对于轻微饱和与严重饱和都适用,且饱和越严重判据识别的灵敏度越高,有利于提高差动保护的正确动作率。EMTP仿真验证了新原理的有效性。
A novel current transformer (CT) saturation detecting method based on parameter identification is proposed. By analyzing the fault component network of external faults, it is derived that the voltage and current at the relay location satisfy an equivalent L-R model. When the CT is not saturated, the parameters identified by least square method reflect the real parameters of the system and have barely variation. However, once the CT is saturated, the parameters identified by side current will have great fluctuation. Thus, by calculating the dispersions of parameters fluctuation, CT saturation can be detected. The performance of this proposed method is applicable to the DC component and harmonics, is not affected by operation mode of system and fault inception angle, so the criterion is easy to set. The algorithm of the criterion based on time-domain shows a fast response time and is selective. Moreover, this method is able to detect all types of CT saturations. The more severe the saturation is, the more sensitive the criterion is. It can be used to improve the performance of differential protection. EMTP simulation results verify the effectiveness of the proposed method.