In this paper, an energy coordination control method based on intelligent multi-agent systems (MAS) is proposed for energy management and voltage control of a DC microgrid. The structure of the DC microgrid is designed to realize the mathematical modeling of photovoltaic cells, fuel cells and batteries. A two-layer intelligent MAS is designed for energy coordination control: grid-connection and islanding of a DC microgrid is combined with energy management of PV cells, fuel cells, loads and batteries. In the hidden layer and the output layer of the proposed neural network there are 17 and 8 neurons, respectively, and the “logsig” activation function is used for the neurons in the network. Eight kinds of feature quantities and 13 different actions are taken as the input and output parameters of the neural network from the micro-source and the load, and the as the control center agent’s decision-makers. The feasibility of the proposed intelligent multi-agent energy coordination control strategy is verified by MATLAB/Simulink simulation, and three types of examples are analyzed after increasing the load. The simulation results show that the proposed scheme exhibits better performance than the traditional approaches.
In a PV-dominant DC microgrid, the traditional energy distribution method based on the droop control method has problems such as output voltage drop, insufficient power distribution accuracy, etc. Meanwhile, different battery energy storage units usually have different parameters when the system is running. Therefore, this paper proposes an improved control method that introduces a reference current correction factor, and a weighted calculation method for load power distribution based on the parameters of battery energy storage units is proposed to achieve weighted allocation of load power. In addition, considering the variation of bus voltage at the time of load mutation, voltage secondary control is added to realize dynamic adjustment of DC bus voltage fluctuation. The proposed method can achieve balance and stable operation of energy storage units. The simulation results verified the effectiveness and stability of the proposed control strategy.