DC microgrid power flow optimization by multi-layer supervision control. Design and experimental validation

Highlights: • DC microgrid (PV array, storage, power grid connection, DC load) with multi-layer supervision control. • Power balancing following power flow optimization while providing interface for smart grid communication. • Optimization under constraints: storage capability, grid power limitations, grid time-of-use pricing. • Experimental validation of DC microgrid power flow optimization by multi-layer supervision control. • DC microgrid able to perform peak shaving, to avoid undesired injection, and to make full use of locally energy. - Abstract: Urban areas have great potential for photovoltaic (PV) generation, however, direct PV power injection has limitations for high level PV penetration. It induces additional regulations in grid power balancing because of lacking abilities of responding to grid issues such as reducing grid peak consumption or avoiding undesired injections. The smart grid implementation, which is designed to meet these requirements, is facilitated by microgrids development. This paper presents a DC microgrid (PV array, storage, power grid connection, DC load) with multi-layer supervision control which handles instantaneous power balancing following the power flow optimization while providing interface for smart grid communication. The optimization takes into account forecast of PV power production and load power demand, while satisfying constraints such as storage capability, grid power limitations, grid time-of-use pricing and grid peak hour. Optimization, whose efficiency is related to the prediction accuracy, is carried out by mixed integer linear programming. Experimental results show that the proposed microgrid structure is able to control the power flow at near optimum cost and ensures self-correcting capability. It can respond to issues of performing peak shaving, avoiding undesired injection, and making full use of locally produced energy with respect to rigid element constraints