Real-time overvoltage prevention control via multi-agent based community energy management systems

High degree penetration of residential rooftop Photovoltaic (PV) systems will cause the overvoltage problem in power distribution system. In this study, a real-time overvoltage prevention control realized by multi-agent based community energy management system (CEMS) is proposed. The main concept of the proposed control is to suppress the overvoltage by managing the operation schedule of household loads. The scheduling of each CEMS is formulated as a combinatorial nonlinear time-series scheduling optimization problem in which the profit of each community is maximized. The effectiveness of proposed method is validated in a 1800-customer distribution system model.