Robust decentralized control for effective load sharing and bus voltage regulation of DC microgrid based on optimal droop parameters

Abstract

The droop control approach is widely used in DC microgrids. The conventional droop control cannot achieve equal current sharing in per unit as well as desired bus voltage regulation. The load sharing error is raised when the cable line parameters of the parallel source converter are unequal. The droop parameter (droop gain and nominal voltage reference) of source converters of the loading condition is the key parameter for effective load sharing and bus voltage regulation. In this paper, a robust decentralized control scheme is implemented, which is based on the optimal droop parameter for mitigating the aforementioned problem without any communication with other control modules. The droop parameters of the proposed control scheme are optimized for the different loading conditions with the help of the hybrid Particle Swarm–Nelder–Mead optimization algorithm. The hysteresis current controller is utilized in place of the cascaded proportional integral controller because tuning of these controllers is not easy. The stability analysis of the proposed system is assured for a specific range of droop gain. The performance of the proposed decentralized control scheme for three parallel source converters and common load is verified in the MATLAB/Simulink environment and validated with a Hardware-in-the-loop real-time simulator based on the dSPACE 1202 platform.