基于数学形态学的电流互感器饱和识别判据

利用数学形态学滤波器结合传统的时差法，实现了一种适用于差动保护的电流互感器饱和闭锁方案。由于数学形态学滤波器具有极佳的奇异点识别能力和噪声抑制能力，使得故障的发生和产生差流浪涌之间的时间差能通过多分辨率形态梯度进行实时、高精度的提取。因此，在TA饱和的情况下，通过合理设定该时问差的门槛值，外部故障和任何内部故障均可做到明显的区分。RTDS仿真结果表明，该方案可有效地防止因大穿越电流引起的电流互感器饱和而造成的差动保护误动，同时可保证对于内部故障的快速反应。

A NOVEL METHOD TO IDENTIFY THE SATURATION OF THE CURRENT TRANSFORMER USING MATHEMATICAL MORPHOLOGY

A new TA saturation blocking scheme used for differential protection implemented with mathematical morphology and the method to locate the time difference between fault occurrence and the differential current emerge is presented. On basis of the excellent singularity detecting together with the noise-repressing abilities of the multi- resolution morphological gradient, the foregoing time difference can be extracted in real time with very high accuracy. Therefore, this time difference can be utilized to identify the external fault from any internal fault in the case of TA saturation. The RTDS based simulation results show that the scheme can effectively prevent the differential protection from mal-operating. Meanwhile it ensures the high-speed response to the internal fault.