Abstract: In DC microgrid systems, power quality issues can cause bus voltage fluctuations and affect the control performance of the system. To achieve a system performance improvement, this paper proposes a performance improvement control strategy based on model predictive control and residual generators for Buck converter parallel systems. First, a state-space model of a multi-Buck converter parallel system is developed. Secondly, the impact of power quality problems on the control performance of the system is analyzed based on droop control and current predictive control. Thirdly, based on the robust control theory, the performance improvement control structure based on the residual generator is derived. And the improvement of the current prediction control cost function and the design of the compensation controller is completed. Then, a least-squares-based parameter identification algorithm is designed for the effect of converter filter inductor and capacitor parameter variations on the control. In addition, a large-signal stability analysis of the system is carried out using mixed potential theory. Finally, experiments are designed on a PXI Target hardware-in-the-Loop simulation experiment platform to verify the effectiveness of the proposed control strategy.