A Dynamic Diffusion Algorithm for Distributed Secondary Control of DC Microgrids

This paper proposes a dynamic diffusion algorithm based distributed secondary control method in DC microgrids (MGs). Rather than conventional consensus algorithm, distributed communication between neighboring converters is implemented using a diffusion algorithm. The proposed approach improves the accuracy of power sharing in the absence of a central controller. In order to accurately represent the system behavior, a discrete-time model is established considering the digital implementation of controllers and communication network. The influence of communication rate, the observation weight of diffusion as well as secondary control parameters on the system stability are investigated using the developed z-domain model of the whole MG. Case studies in MATLAB/Simulink verify the performance of the proposed method under load changes, link failure and converter faults. It has been demonstrated that the proposed dynamic diffusion algorithm shows the advantages over consensus in the aspect of convergence speed.