Analysis Of Low Frequency Oscillations Based On Wide Area Measurement System And Improved Prony Algorithm

Power system stability in our country is easier to be undermined because of interconnected network speeding up, electric power market introduced, generation unit capacity increased as well as large-capacity energy transmission pattern shaping, and low frequency oscillation which occur frequently is one of the major problems of power system stability. This article presents the analysis of low frequency oscillation in power system in improved Prony algorithm based on wide area measurement system.WAMS is discussed in the structures, features as well as applications in power system, and wide area measurement signals are prepare for analysis of low frequency oscillation in power system because of WAMS can grasp the dynamic information from signals characterized by synchronization and wide-measurement.Prony algorithm widely used in power system is researched in both reckon of sample function matrix order in two criterions and algorithm parameters simplified calculation. To improve Prony algorithm, similar, non-similar signals are defined and sorted multi-signals improved Prony algorithm is proposed to analyze low frequency oscillation.In large-scale system, the paper makes a further effort to improve the Prony algorithm. Model energy correlativity is defined from Prony algorithm parameters and structured into correlation matrix based on correlation factor matrix in small signal stability analysis. The correlation matrix can reflect generation unit correlativity and be as basis of dividing signals in sorted multi-signals improved Prony algorithm.The proposed improved Prony algorithms were applied in EPRI-36 system to analyze low frequency oscillation, and the accuracy as well as the validity is demonstrated.