Online Electromechanical Mode Identification from Power System Ringdown Responses

This paper introduces a comprehensive methodology for the online identification of electromechanical modes. Within the proposed framework, a series of tasks is executed prior to mode estimation in order to efficiently handle ringdown data. These tasks include detecting the onset of a disturbance, extracting the optimum length of the ringdown, and further pre-processing the signal for enhanced analysis. To perform mode identification the Hybrid frequency/time-domain approach is used. The performance of the proposed methodology is tested using a synthetic signal which is intentionally distorted with noise and contains non-typical data. Results indicate the applicability and efficiency of the proposed method for the online identification of power systems electromechanical modes.