Controlled islanding, which splits the whole power system into islands, is an effective strategy against rolling blackout during severe disturbances. Finding the islanding solutions in a real-time manner is complicated because of the combinatorial explosion of the solution space occurs for a large power network. In this work, a computationally efficient controlled islanding algorithm is proposed that uses constrained spectral clustering while addressing the generator coherency problem. The objective function used in this controlled islanding algorithm is the minimal power-flow disruption. The sole constraint applied to this solution is related to generator coherency. An undirected edge-weighted graph is created based on absolute values of apparent power flow and constraints related to transmission line availability and coherent generator groups are included by altering the edge weights of the graph and using a subspace projection. Spectral clustering is then applied to the constrained solution subspace to find the islanding solution. The methodology is tested on an IEEE-39 test system with a fully dynamic model. Simulation results demonstrate the efficacy of our approach.

中文翻译：

---

基于约束聚类的电网孤岛方案

控制孤岛将整个电力系统分裂成孤岛，是在严重扰动期间防止滚动停电的有效策略。由于大型电网会发生解空间的组合爆炸，因此以实时方式寻找孤岛解很复杂。在这项工作中，提出了一种计算效率高的受控孤岛算法，该算法在解决生成器一致性问题的同时使用受约束的谱聚类。该受控孤岛算法中使用的目标函数是最小的潮流中断。应用于此解决方案的唯一约束与生成器一致性有关。无向边加权图是基于视在潮流的绝对值创建的，并且通过改变图的边权重和使用子空间投影包括与输电线路可用性和相干发电机组相关的约束。然后将谱聚类应用于约束解子空间以找到孤岛解。该方法在具有完全动态模型的 IEEE-39 测试系统上进行了测试。仿真结果证明了我们方法的有效性。