Graph Partitioning Method for Distributed and Parallel Optimal Power Flow

The power network partition is an important issue for the auxiliary problem principle based distributed and parallel optimization algorithm (APPBDPOA) as it affects the convergent performance of the algorithm. A maximum-cardinality matching based fast graph partitioning method (MCMBFGPM) is proposed. After the power network is initially partitioned into several sub-areas, this method is adopted to reduce the numbers of frontier buses and balance the numbers of interior buses of all sub-areas. The numerical results show that it converges fast and can significantly speed up the auxiliary problem principle based distributed and parallel optimization process