An online energy management method is proposed for DC nanogrid cluster to control the energy flow within each nanogrid and among the nanogrids. Based on the Lyapunov optimization theory

the concept of virtual energy storage queues is developed. A mathematical model is proposed to improve the solar replacement ratio and guarantee queue stability

which decouples the complex energy management problem into linear programming problems of each time slot. Only the current information is needed to make the real-time decision of the exchange electricity among nanogrids. Simulation is conducted based on the actual household load and photovoltaic generation data collected. The results have verified that the proposed method can effectively improve the accommodation of renewable energy in the DC nanogrid cluster while satisfying the load demand. Meanwhile

the test result shows that it does not take the method long to make decisions

which is the real-time requirement of the system.