A Novel Fast Searching Algorithm for Power System Self-Adaptive Islanding

Self-adaptive system islanding refers to the separation of an interconnected power system into electrically isolated islands automatically based on the identification of unstable mode. A fast searching algorithm for self-adaptive power system islanding, which is called multilevel reduced graph partitioning (MLRGP) algorithm, is presented in this paper. The aim of this algorithm is that the generation load imbalance in each island is as small as possible. The algorithm computes a reasonable islanding strategy of power system G=(V, E) in O(|E|) time. A C++ program is developed based on the algorithm. The verification of the islanding program is proven with simulations on a practical 212-bus power system. It costs only about 20 ms to get a reasonable islanding strategy on an ordinary PC, which satisfies the speed demand of self-adaptive system islanding.