A Dual‐Loop Coordinated Control Strategy for PV‐Storage VSG With Cascaded Model Predictive Control and Adaptive Inertia‐Damping Mechanism

ABSTRACT Conventional photovoltaic‐storage virtual synchronous generators (VSG) often suffer from active power overshoot, frequency oscillations, and limited stability during grid‐connected operation. This paper proposes a dual‐loop coordinated optimization strategy. First, in the outer loop, a radial basis function (RBF) neural network adaptively tunes virtual inertia and damping, while a staged variable‐weight model predictive control (MPC) introduces real‐time power compensation to suppress frequency deviations and enhance dynamic recovery. Second, in the inner loop, a delay‐compensated finite control set MPC (FCS‐MPC) selects optimal switching vectors to achieve accurate current tracking and improve power quality. Simulations show that, compared with conventional MPC‐VSG, the proposed method reduces the maximum frequency deviation from 0.445 Hz to 0.297 Hz, shortens the recovery time from 0.393 to 0.164 s, eliminates active power overshoot from 16.2% to 0%, and decreases the current total harmonic distortion (THD) from 5.53% to 2.97%, achieving a 46.3% reduction. These results confirm that the strategy effectively suppresses frequency and power fluctuations, accelerates transient recovery, and strengthens grid‐connected stability, demonstrating strong potential for engineering applications.