This article concerns the assessment of diagnosis performance of a recently developed secondary protection (SP) applied to the IEEE 16‐machine 68‐bus test system subject to primary protection (PP) misoperations on transmission short circuit faults. A PP may fail to trip or falsely trip a line. Such PP misoperations have been known to be a main culprit of modern day blackouts. The function of the SP is to take a timely corrective action before a misoperation causes instability through on‐line diagnosis and stability assessment. Its feasibility hinges on the effective utilization of networked advanced digital relays. We have recently built a full set of diagnosis filters with sensor‐localized dynamic transmission circuit models representative of all anticipated line contingencies in the 68‐bus system and computed the postfault stability regions for the contingencies. The sensor‐localized circuit partition drastically reduces the processing and communication complexity of diagnosis. This work evaluates the diagnosis performance of the SP's multiple‐model filters (MMF) using hybrid simulations driven by both randomly generated discrete events of PP misoperations and the governing dynamics of rotor angles of the 16 interconnected synchronous generators. Of the 1000 independent replications of simulation conditioned on each type of randomly generated misoperations, a 98% correct diagnosis rate is achieved. Each diagnosis conclusion is reached by the SP's MMF scheme within a few tens of a millisecond from the onset of a PP misoperation following a random transmission short circuit fault. This suggests that the SP can circumvent a large percentage of potential blackouts cost‐effectively.

中文翻译：

---

68总线电源系统二次保护的诊断性能评估

本文涉及评估最近开发的二次保护（SP）的诊断性能，该二次保护适用于IEEE 16机68总线测试系统，该系统在传输短路故障时遭受主要保护（PP）错误操作。PP可能无法使线路跳闸或误跳闸。众所周知，这种PP误操作是现代停电的主要原因。SP的功能是通过在线诊断和稳定性评估，在误操作导致不稳定之前及时采取纠正措施。它的可行性取决于网络高级数字继电器的有效利用。最近，我们用传感器定位的动态传输电路模型构建了一套完整的诊断滤波器，这些模型代表了68总线系统中所有预期的线路突发事件，并计算了突发事件后的稳定区域。传感器本地化的电路分区大大降低了诊断的处理和通信复杂性。这项工作使用混合模拟来评估SP的多模型滤波器（MMF）的诊断性能，该混合模拟是由随机产生的PP错误操作的离散事件以及16个互连的同步发电机的转子角控制动态所驱动的。以每种随机产生的错误操作为条件的1000个独立模拟仿真中，可实现98％的正确诊断率。SP'得出每个诊断结论 在随机传输短路故障之后的PP错误操作开始后的几十毫秒内使用MMF方案。这表明SP可以经济高效地规避很大一部分潜在的停电情况。