Inertia Identification for Grid-forming Converter Based on Model-Data Hybrid-Driven Method

The integration of grid-forming converters (GFMC) into the power grid significantly enhances the frequency stability of renewable energy systems. In this context, accurately evaluating the inertia parameters of converters is crucial for assessing the frequency regulation potential of power systems. However, inertia identification techniques that rely solely on parameter-based models often produce significant errors due to their inability to account for nonlinear factors. To overcome this limitation, this paper proposes an inertia parameters identification technology based on model-data hybrid-driven method for GFMC. Firstly, the synchronous power parameter model and nonlinear components of GFMC are analyzed. To address the nonlinear components in the inertia model, a data-driven fitting technique utilizing a multilayer perceptron is employed, enabling precise estimation of inertia parameters. Finally, numerical simulations are conducted to validate the effectiveness and accuracy of the proposed method.