An optimization of distribution system with Distributed Generators (DGs) is a big challenge in recent power systems. In our previous study(23) a real-time control method has been proposed to track an optimal operating point that is the common optimum for the global problem to maximize the total generations of DGs and for the local individual problems to maximize the profits of local DG generations. By extending the previous study, this paper proposes an advanced method to improve the robustness of the control. Instead of solving the local optimizations in the individual DGs, the target nodal voltages and P&Q prices are provided as the individual control targets. Based on multi-agent architecture, an effective distributed control is performed effectively. A management agent performs an optimal power flow computation to determine the optimal voltage profiles and the shadow prices of real and reactive powers, which are transferred to the local individual agents through a common blackboard memory. The proposed method is effective to mitigate the voltage violation problem and maximize the total PV power generations. The method also resolves a present problem that some specific PVs tend to be deactivated due to voltage violations.
J-STAGEがリニューアルされました!https://www.jstage.jst.go.jp/browse/-char/ja/