An application of the successive shortest path algorithm to manage power in multi-agent system based active networks

In order to handle a high penetration of distributed generation (DG), the existing passive distribution systems need to become more active. In the so called active network (AN), power flow management has developed to cope with bidirectional power flow and the intermittence of DGs. This paper proposes a method to better manage active power flows based on graph theory application and multi-agent system (MAS). While the power system is represented as a directed graph, the optimal power flow (OPF) is understood as a minimum cost flow problem that can be solved by the successive shortest path algorithm. The algorithm is implemented in a distributed way and is supported by MAS. Simulations using a five sub-areas network show that the method is able to reach the optimal operation state while restraining power flows within defined bounds. Furthermore, the method can also response changes in power generation cost, network configuration and load demand increase. Copyright (C) 2009 John Wiley & Sons, Ltd