In modern power systems, the extensive integration of renewable energy sources leads to a reduction in system inertia, thereby affecting grid stability. This study proposes a novel method for estimating grid inertia, which, under quasi-steady state conditions, effectively incorporates load contribution by combining class noise signals collected by PMUs (Phasor Measurement Units) with the Forgetting Factor Unscented Kalman Filter algorithm (FF-UKF). The research initially analyzes the characteristics of various inertia sources in power systems dominated by renewable energy. Building on this, we propose a grid node equivalent inertia estimation model based on node equivalent frequency response, developing an FF-UKF-based algorithm that allows for the continuous dynamic estimation of inertia at various grid nodes. Simulation experiments conducted within the IEEE 39 bus system validate the method’s effectiveness and demonstrate the spatiotemporal distribution of grid inertia under varying levels of renewable energy penetration. This study is significant for understanding and enhancing frequency control and stability in power systems, providing key technical support for the modernization of the power system and the efficient integration of renewable energy.