A distributed control strategy (DCS) is an efficient way to realize power sharing among distributed generators (DGs) and restore the voltage deviation caused by droop control. However, there are several limitations in the traditional strategy of hybrid ac/dc microgrid (MG), such as the heavy communication burden and sensitivity to the working conditions. In this article, a distributed model-free adaptive control strategy (DMFAC) is proposed with an event-triggered (ET) mechanism. A unified control framework of hybrid MG is established and the ET-DMFAC can operate within it to realize multiple control strategies simultaneously. The dynamic linearization parameters of hybrid MG are estimated online to reflect the working condition, and an ET mechanism is designed to reduce the communication burden. The stability of ET-DMFAC is analyzed with the discrete Lyapunov theory and experiments are carried out to verify the effectiveness of the proposed ET-DMFAC strategy under different operating conditions. The outperformance of ET-DMFAC is demonstrated by comparing it with many existing DCS of hybrid AC/DC MG.
24 October 2024 in IEEE Transactions on Energy Conversion
Parallel interlinking converters (ICs) find application in high-power hybrid AC/DC microgrids. However, challenges exist in achieving coordinated control among different ICs for power sharing, which is beyond this parallel structure and encompasses the interplay between different subgrids. The existing approach suffers from a heavy communication burden and sensitivity to disturbances. To address this issue, this paper proposes a distributed robust sliding-mode control (SMC) for the hybrid AC/DC microgrid with parallel ICs, ensuring the convergence of multiple control objectives within a predetermined time slot. An observer is designed to mitigate disturbances for a reduction in the required controller gain and suppression of chattering issues. The stability of the proposed control strategy is analyzed using the Lyapunov theory and control parameters are designed based on the stability analysis. Experiments under various working conditions demonstrate the outperformance of the proposed method compared to traditional distributed control approaches.