Prony Analysis for Extracting System Oscillation Mode Characteristics

In this paper, we explore the application of Prony analysis, which serves as a powerful technique for characterizing system oscillation modes. By analyzing system responses to given input signals, Prony analysis enables direct estimation of oscillation frequency, damping ratio, amplitude, and initial phase parameters. The implementation typically involves solving a linear prediction model through eigenvalue decomposition of the system matrix, where the Prony algorithm fits damped complex exponentials to uniformly sampled time-domain data. However, in power systems, large-scale disturbances can dominantly influence low-frequency oscillation modes, potentially compromising analysis accuracy. To address this challenge, we propose a novel methodology based on the Prony algorithm that discriminates dominant low-frequency oscillation modes under significant disturbances through energy-level analysis of oscillation patterns. Our research details the implementation principles of this approach, including the mathematical formulation for calculating mode energy distributions and threshold-based identification criteria. The method's advantages include improved mode selectivity and disturbance resilience through energy-weighted parameter evaluation. Experimental validation using power system simulation data demonstrates the method's effectiveness in identifying critical oscillation modes under various disturbance scenarios.