Distributed Primal-Dual Method for Current Sharing and Voltage Regulation in DC Microgrids

Due to the strong coupling between power injection and bus voltage in the DC microgrid, it is challenging to achieve both economic dispatch and restoration of nominal voltages for all buses simultaneously. To address this problem, this paper first proposes a control model based on bi-objective optimization to balances economic efficiency and voltage distribution quality. Furthermore, a distributed weighted gradient primal-dual method is introduced, as even short-term mismatches between generation and demand in the system can lead to voltage collapse. Specifically, this method obtains the gradient information with aggregated variables in a fully distributed manner by the collaborative transmission of weighted information between agents. Compared to existed methods, this method not only maintains supply-demand balance at all times, but also dynamically tracks the optimal operating point, thus achieving online distributed optimization. Finally, the stability and convergence of this method are theoretically analysed and its effectiveness in real-time energy management is validated by simulation studies.