Abstract: As a key demonstration project for high-proportion renewable energy integration in China, the desert-gobi-barren land large-scale renewable energy base exhibits a complex system architecture with strong inter-station coupling. The dynamic variation of grid strength and impedance characteristics poses substantial challenges to system stability assessment and operational security. Conventional impedance identification methods, which typically rely on local disturbance injection or single-point measurements, are unable to accurately capture the interactions among multiple stations within a unified grid network. To address these challenges, a system-level equivalent impedance monitoring method based on a unified grid framework is proposed. By formulating an equivalent impedance identification model for renewable generation units, the proposed method uniformly represents the interactive effects between the grid and multiple stations as system-level impedance parameters. Dynamic impedance estimation is achieved using an improved forgetting-factor recursive least squares(FFRLS) algorithm, enabling real-time evaluation of grid strength. Compared with traditional identification approaches, the proposed method provides an integrated equivalence of multi-station interactions in complex grid structures, eliminates the need for external disturbance injection, and enhances the safety and real-time performance of impedance identification. Furthermore, the introduction of a forgetting factor improves the algorithm’s adaptability to time-varying grid characteristics. Simulation and experimental results validate the accuracy, responsiveness, and overall effectiveness of the proposed method.