Multi-Scenario Parameter Estimation for Synchronous Generation Systems

Parameter estimation of synchronous generation systems is vital to the validation of power system dynamic analysis. Compared with the substantially investigated single-scenario parameter estimation, multi-scenario estimation is more advantageous in terms of estimation accuracy. This study proposes a systematic method for solving the challenges of implementing multi-scenario parameter estimation through four crucial steps. First, a novel method using trajectory sensitivity is proposed for multi-scenario parameter identifiability analysis. Scenarios are then ranked based on the scenario identifiability index with respect to the identifiable parameters. Third, a scenario decomposition strategy is developed by using the reduced-space interior point method to accelerate the parameter estimation procedure. Finally, a method based on Chi-square test is proposed for bad scenario detection and identification. In the case study, numerical experiments and field measurements are used to validate the effectiveness of the proposed method.