Modal analysis of active distribution networks using system identification techniques

Mode identification from post-disturbance “ringdown” responses can provide vital information concerning the dynamic performance and the stability margins of power systems. Therefore, several measurement-based identification techniques have been proposed in the literature to analyze ringdown responses of transmission systems and provide close to real-time estimation of the modal content. However, the applicability of these methods has not been thoroughly investigated for the analysis of active distribution networks (ADNs). Scope of this paper is to evaluate the applicability and the performance of eight measurement-based system identification techniques for the modal analysis of ADNs. The examined methods are used to identify the dominant oscillatory modes contained in ringdown responses of different types of signals. The Monte Carlo method is applied to investigate the influence of several parameters on the accuracy and efficiency of the identification procedure, while laboratory measurements are used to further demonstrate the accuracy of the examined methods. Practical issues encountered in the application of the identification techniques for the analysis of ADNs are discussed and potential solutions are proposed. Results reveal that although most of the examined techniques perform satisfactorily enough and thus can be readily employed for the modal analysis of ADNs, the Vector Fitting and the Hybrid FD/TD seem to be the most effective methods in terms of accuracy, robustness and computational efficiency.