To address the problem of poor voltage stability of wind-solar-energy storage DC microgrids in distributed renewable energy systems
an optimized nonlinear active disturbance rejection control strategy based on the SAC algorithm (SAC-ADRC) is proposed. Firstly
the wind-solar-energy storage system is modeled with nonlinear ADRC control. Then
the gain parameters of the nonlinear ADRC are reconstructed by using linear/nonlinear ADRC switching to improve its internal parameters which are more difficult to tune and analyze. Finally
the analysis establishes a mechanism for SAC intelligence to learn interactively with the microgrid environment
enabling the adjustment of non-linear ADRC parameters. Comparative analysis using algorithm convergence curves and simulation of various classical working conditions confirms the superiority of the SAC-ADRC control strategy in terms of interference performance. Thus
it is shown that the organic integration of nonlinear ADRC and deep reinforcement learning improves the stability of the microgrid bus voltage.