提出了适用于电力系统低频振荡模态识别的改进多信号矩阵束算法。利用奇异值分解（Singular value decomposition，SVD）分离信号和噪声子空间，确定阶教并消除信号噪声。通过建立多信号归一化的样本函数矩阵对矩阵束算法进行改进，辨识电力系统模态。利用原始Prony法、谐波恢复的Prony法和改进的多信号矩阵束法，对理想信号和仿真系统进行分析。结果表明多信号矩阵束法的辨识精度较高，具有一定的抗噪能力，并且通过对多信号归一化的处理避免了不同类型信号叠加时较小信号的湮没，适用于低频振荡在线识别。

Modern power system has become complex and the size of the nonlinear systems makes the traditional analysis method present more and more limitations. Therefore this paper introduces an improved multi-signal matrix pencil algorithm suitable for low-frequency oscillations modal identification of power system. First, the method of singular value decomposition (SVD) is used to separate the signal from the noise sub-space, confirm the order and eliminate the signal noise. And then through the establishment of multi-signal normalized experiment matrix, the matrix pencil algorithm is improved to identify power system mode. The ideal signal and simulation system are analyzed by using the original Prony method, the harmonic retrieval Prony method and the matrix pencil method. Results show that multi-signal matrix pencil method has high precision and a certain anti-noise ability. Through the normalized processing of the multi-signal, the submerging of small signals is avoided when different type

电压波动与闪变是当前电能质量问题的重要研究对象之一,对其进行有效监测,对于电能质量治理具有重要意义。提出一种基于Hilbert变换与改进的Prony法的参数估计法,该方法能准确提取出信号的包络线,检测出调制分量的幅值、频率等参数信息,采用统计排序法计算出短时闪变值Pst。MATLAB仿真验证了该算法的有效性与精确性。

Voltage fluctuation and flicker are currently important research subjects in power quality. Effective monitoring of the two fac-tors is of great significance to ensuring power quality. An evaluating method based on Hilbert transform and improved Prony method is proposed to evaluate the modulating wave. The method can precisely detect the range of the voltage fluctuation as well as the modulation amplitude,frequency and other parameters. The short term flicker value Pst are calculated by the statistical ranking method. The simula-tion data by MATLAB show the accuracy of the theory.

Prony算法和Hilbert-Huang变换（HHT）算法是目前实际应用于低频振荡信号辨识的两种算法。Prony算法对于无噪声干扰的平稳信号的分辨率和准确性较高。HHT算法具有时频分析和滤除高频分量的能力，抗噪声性能好，计算速度快，并可以分析主导模式随时间改变的非平稳信号。从准确性、稳定性、完备性、快速性四个方面比较了两种算法在低频振荡在线辨识中的适用性，指出增加前置滤波环节的Prony算法可以满足低频振荡在线监测的要求，而实现强迫振荡扰动源定位则必须采用可进行时频分析的HHT算法。通过实际算例验证了两种算法的适用性。

Prony and Hilbert-Huang Transform (HHT) are two signal processing techniques utilized for on-line low-frequency oscillation detection. Prony algorithm is appropriate for processing stationary signals without noise. HHT algorithm has the ability of filtering high-frequency noise and processing non-stationary signals. The performance of Prony and HHT algorithms is compared from following aspects: accuracy, stability, completeness and speediness. Prony algorithm which includes the pre-filtering can meet the requirements of low-frequency oscillation monitoring, while HHT can be used for identification of disturbance source. The applicability of the two algorithms is validated with real power system data.

研究低频振荡特征辨识算法是实现电力系统低频振荡监视的重要理论基础。文章基于Prony算法,结合变量投影（VP）,采用奇异值分解（SVD）,应用可分离最小二乘,辨识出Prony算法数学模型参数,不仅避免了传统Prony算法矩阵求逆困难和误差较大的问题,而且降低了参数空间的维数,简化了搜索空间的拓扑结构,增强了模式识别的抗噪性能和运算速度。仿真结果表明,该算法提高了电力系统低频振荡模式识别的效率和精度。

Low frequency oscillation characteristic analysis algorithms are used to monitor power sys-tem low frequency oscillation for the control of wide-area damping .In this paper ,the Prony method is integrated with variable projection (VP) approach ,the singular value decomposition (SVD) is used , and the separable nonlinear least squares are applied to identify the parameters of Prony mathematical model .The presented method not only avoids the problem in matrix inverse and larger error of the traditional Prony method ,but also reduces the dimension of the parameter space and simplifies the to-pology structure of the search space .It enhances the antinoise performance of mode identification and improves the computing speed .The simulation results show that the proposed method improves the accuracy and efficiency of the power system low frequency oscillation mode identification .

针对Prony算法抑噪能力差、计算效率低的问题,分析Prony算法的原理和主要参数的选择策略,提出一种基于Prony算法的发电机组间功率低频振荡在线辨识新方法,从模型有效阶数确定、AR参数估计、数据预处理等方面说明该方法的计算步骤,通过算例仿真及结果分析,认为该方法能够满足电力系统低频振荡在线辨识的需要。

As the traditional algorithm is sensitive to noise and has ineffective calculation ,the basic principle of Prony algo-rithm is introduced in this paper ,a new Prony method for on-line analysis of low-frequency oscillations between power gen-erators is discussed in this paper .The calculation steps of the method are described by way of the identification of the practi-cal order of the model ,estimation of the parameters and the preprocess of the signal .The simulation results show that this method can meet the needs of low-frequency oscillations online identification .