The high penetration of renewable energy has led to the widespread emergence of weak grids, under which grid-following (GFL) inverters suffer from dynamic instability induced by phase-locked loops (PLL). Although reducing the bandwidth of the PLL can enhance system stability, it significantly degrades the ability of the inverter to reject power interaction disturbances. The inherent trade-off between the bandwidth of the PLL and its disturbance suppression capability are investigated. A cooperative strategy through a grid-forming (GFM) inverter to improve the robustness of GFL inverters without requiring hardware modifications is introduced. The simulation results validate that the proposed strategy effectively suppresses power interaction disturbances and enhances system stability under weak-grid scenarios.

The integration of photovoltaic (PV) systems into weak-grid environments presents unique challenges to the stability of grid-connected inverters. This review provides a comprehensive overview of the research efforts focused on investigating the stability of PV grid-connected inverters that operate under weak grid conditions. Weak grids are characterized by a low short-circuit capacity and low inertia, making it essential to explore strategies that enhance the stability and performance of inverters in such challenging environments. This review covers various aspects, including control strategies and advanced technologies implemented to address stability problems. The research findings related to the impact of weak grid conditions on PV inverters, modeling techniques, and analysis results are discussed. Additionally, this review highlights emerging trends, identifies gaps in the current research, and suggests potential avenues for future investigations aimed at improving the stability of PV grid-connected inverters in weak grid scenarios.

Equivalent models are essential for time-domain simulation and harmonic stability evaluation of renewable power plants. Such harmonic instability may co-occur at the plant-level and unit-level, or it may only occur at the unit-level, manifesting as local resonances. However, conventional equivalent methods can only simulate plant-level instability, but cannot consider such unit-level instability. To address this problem, the main contributions of this paper are as follows: 1) Based on the impedance matrix and the eigenvalue analysis, the reason why conventional equivalent methods cannot accurately simulate uni-tlevel stability is studied. 2) A double-machine equivalent method is proposed. The model used in this method is a minimumscale equivalent model capable of simulating plant-level and unit-level stability, simultaneously. It uses two single-machine models to represent a generation cluster containing N generation units. A single-machine model M₁ represents N - 1 generation units for simplification, while another single-machine model M₂ represents the remaining one generation unit for simulating uni-tlevel stability. 3) Furthermore, a practical improvement approach of the double-machine equivalent method is proposed for taking into account operating point differences and complex connection conditions. Finally, the proposed method is verified by simulations and experiments.