Inverter parameters estimation using structured state-space and prediction error method

New power systems incorporate various power electronics technologies, such as Inverter-Based Resources (IBRs), High Voltage Direct Current (HVDC) systems, and inverters. These components create instability issues for network operators, which need to be thoroughly studied. However, limited information about these devices’ physical and control parameters poses challenges in the modelling of those devices. System identification is an approach that can be employed to estimate different parameters for stability assessment purposes. This study proposes a structured state-space approach using the Prediction Error Method (PEM) to estimate the coupling filter parameters (L & C) and the current control gains of a grid-following inverter using a small-signal model. The results demonstrate accurate estimations for these parameters, with errors less than 5% for a strong network. The impact of uncertain control structures is also assessed. In addition, weak network configuration and the impact of it on the estimation. The uncertain control structure impacted the estimation, but it was only observed on the high-frequency poles. The network mismatch case accuracy varies depending on the excitation-signal length which can lead to a high error percentage. The time domain response using the estimated parameters is also investigated and it shows a close response with the actual model. This attempt is to estimate inverter parameters to create more realistic models that can be used for stability assessment for network operators. Additional validation indicates promising results, suggesting further investigation of this approach for inverters can be conducted in the future.