电力系统振荡与谐波扰动识别的prony分析法  

Prony Analysis for Identification of Transient Oscillation and Harmonics

作　　者：易新超[1] 徐永海[1] 肖湘宁[1] 陈磊[1] 

机构地区：[1]华北电力大学电力系统保护与动态安全监控教育部重点实验室,北京102206

出　　处：《现代电力》2005年第5期33-37,共5页Modern Electric Power

基　　金：教育部博士学科点专项科研基金资助项目(20010079002)

摘　　要：为实现电压扰动的抑制与补偿,需要对电压扰动进行自动的识别,以便有针对性地采取措施。针对电压扰动中难以正确区分的谐波与高、低频振荡扰动,提出了一种基于Prony算法的识别方法。该方法能够对谐波和高、低频振荡扰动信号进行有效识别,并求解出相应扰动信号成分的特征参数:幅值、衰减系数、频率和初相位等。对该方法进行了计算机仿真验证,取得了令人满意的效果。信噪比为40dB条件下,谐波、振荡和振荡与谐波扰动同时发生时,在准确计算特征参数情况下的正确识别率分别为96%、94%和92%,若只考虑类型识别正确率则上述指标可达98%。所提方法适用于频率跨度大、成分比较复杂的扰动信号的分析,能够很好实现扰动的识别。将所提方法与其它方法结合,还可对多种扰动进行有效的识别。In order to realize the suppression and compensation of voltage disturbances, voltage disturbances need to be identified properly so that suitable actions can be taken. This paper proposes a new method based on Prony analysis to recognize .the harmonics, high and low frequency oscillation in voltage disturbances, which are normally hard to be identified properly. This method can work out the characteristic parameters such as amplitude, damping factor, frequency, and phase for harmonic components. The proposed method has been implemented and tested. The test results show that under the condition that the SNR （signal noise rate） is 40 dB, the identification rate in the following three cases： pure harmonics, pure transient oscillation, and harmonics and oscillation, are 96%, 94%, 92% respectively, with an assumption that the error in the calculation of parameters used are ignorable. If we are only interested in rec-ognizing the disturbance types, the identification rate can reach 98%. The method is suitable for analysis of complicated disturbance signals with a broad frequency span.

关 键 词：电力系统振荡 普罗尼分析 扰动识别 特征参数 谐波 

分 类 号：TM714[电气工程—电力系统及自动化]