低频振荡预警及分析功能在新疆电网WAMS系统中的应用及改进

对新疆电网广域测量系统高层应用中的低频振荡预警和分析功能进行了深入研究与改进.针对系统中原有的Prony算法分析长数据段时无法辨识出衰减快的分量而导致误差偏大的不足及视窗短的问题,引入时域峰值识别法,将两者有效的结合起来,进行长短视窗分析,有效提高了分析长数据段的准确性.同时建立了两段数据层分析方法及不同系统的数据转化分析方法,将同系统的数据分段、分层进行分析,将不同系统采集到的数据进行对比分析相互验证.通过实例验证了算法的可行性和有效性,对电网运行起到了实际作用,有利于新疆电网低频振荡分析和实时告警,保证了电网的安全稳定运行.     

基于广域信息的电力系统阻尼控制器反馈信号选择

广域阻尼控制器（WADC）是抑制电力系统区间低频振荡模式的一种新技术,区别于传统的本地控制器设计,存在大量的不同种类反馈信号之间如何选择的问题。文中提出以WADC单位控制输出量获得最大阻尼为目标,基于线性系统的理论推导得到了用于WADC广域反馈信号选择的主模比指标,并且给出了基于Prony辨识的可用于复杂系统中的主模比计算方法。在2006年南方电网交直流混联系统中,通过特征值分析和时域仿真验证了最大主模比指标的有效性。     

广域测量系统在吉林电网低频振荡监控中的应用

广域测量系统(W AM S)作为电力系统实时动态检测手段,在复杂的大型电力系统中发挥着重要作用。在介绍了低频振荡在线监视与分析方法的基础上,给出在吉林电网的应用经验,分析了华能长春热电厂2号机组低频振荡产生的原因,并给出有效地应对措施。     

超声信号的双指数模型及参数确定方法

 根据实际系统超声信号的普罗尼分析结果,提出了基于双指数线性组合的超声信号经验模型,分析了上升时间和半峰宽特性,给出了模型参数确定方法.对双指数模型和混合指数模型的拟合精度与TOF估计方差进行了对比分析,结果表明双指数模型更适于描述半峰宽与上升时间比值大于2.5的超声信号.     

基于Prony算法的发电机励磁系统参数辨识

阐述了发电机励磁系统参数辨识的原理。基于Prony算法提出了一种工程上便于应用的发电机励磁系统参数辨识新方法。指出该方法的优点是根据励磁系统输入、输出响应信号直接在时域上实现参数的辨识。在Matlab/Simulink中的仿真结果表明,该算法能准确的辨识出励磁系统各个环节的参数,具有精度好、运算率高、程序通用性强等优点。     

基于Prony算法的次同步谐振检测方法

提出一种可实时检测次同步谐振SSR(sub-synchronous resonance)的方法。基于电气量测量的思想,采样瞬时电流作为检测量。首先对采样电流信号进行滤噪处理,接着利用Prony算法进行分析,辨识出次同步分量的幅值,观察其在连续时间段上的变化规律,进而判定次同步谐振发生与否。此外介绍了Prony算法具体的实现步骤、噪声的处理方法以及Prony算法用于次同步谐振检测的性能优势。最后在IEEE第一标准测试系统模型上进行仿真,试验结果证明了基于Prony算法的次同步谐振检测方法的有效性。     

WAMS-based detection and early-warning of low-frequency oscillations in large-scale power systems

The phasor measurement units (PMU) and wide-area measurement systems (WAMS) have been widely established in modern power systems to improve the monitoring of the system behavior as well as the system control. In this paper, an integrated scheme for the monitoring and detection of low-frequency oscillations has been developed, based on our extensive experience in using Prony algorithm for oscillation analysis in practical WAMS projects. By analyzing the real-time synchro-phasors, the proposed scheme is competent to identify the characteristics of the low-frequency oscillations in real-time and alert the operators once the oscillation is under-damped or divergent. To ensure accurate monitoring of system dynamics and reliable detection of dangerous oscillations with noise-polluted WAMS measurements, several key signal-processing techniques are implemented, including delicate designing of prefilters, Prony analysis result correction, and comprehensive indices for stability evaluation. In the end, the developed scheme is tested with simulated signals and measurements from practical power systems. Its applications to large-scale power grids for detection and early-warning of low-frequency oscillations are also presented. The results have convincingly demonstrated the validity and practicability of the developed scheme.     

基于改进的Prony算法在线辨识电力系统低频排振荡模式

基于传统的Prony算法对输入信号要求较高,同时对分析数据的噪声非常敏感,提出了一种改进的Prony算法,对在线获取的信号进行快速拟合,从而分析出信号的振幅、阻尼比、频率和相角等信息.改进的Prony算法的拟和精度在36节点的多机系统中进行验证,该算法输入信号是基于广域测量系统提供的各机组功角变量.仿真计算结果表明,该改进算法可实现低频振荡主导模式的在线辨识.     

瞬时电压闪变监测及定量评估方法研究

简要介绍了目前电能质量监测中电压波动和闪变的测量方法,并分析了存在的一些问题。提出了一种瞬时电压闪变监测及定量评估方法,该方法首先采用Prony方法将电压信号分解为基频和间谐波频率两部分,然后借助Hilbert变换从Prony线性拟合结果中提取电压包络线,即而获得电能质量评估中瞬时电压闪变严重程度的定量信息。数值仿真算例初步验证了该方法的有效性。     

The Huynen-Fork Polarization Parameters in the Classification of Dielectric Mine-like Objects

We present a polarimetric approach that can be used to characterize subsurface targets by means of ground-penetrating radar. Several quantities related to the Huynen-fork parameters are basic to the study. These quantities are all expressible in terms of the elements of Sinclair matrix S which in turn, can be determined as functions of frequency by an application of the method-of-moment (MoM)—which is briefly reviewed here—to a mine-like object buried at various depths in a soil of specified dielectric and moisture properties. The quantities in question are the Stokes parameters, the elevation, azimuth, and polarizability angles, as well as the polarization ratios and the scattering eigenvalues which optimize the backscattered power. Some of these quantities are studied here and displayed in various graphs. These frequency dependent graphs exhibit useful symmetry properties. The plots of the polarizability angle (β) vs. frequency seem to remain invariant with target depth, and exhibit resonance features that are later shown to agree well with other standard methods to estimate resonances such as Prony method. A simple analysis of the early-time resonances thus estimated is used to obtain acceptable mine dimensions. The agreement found implies that polarimetric techniques, such as present one, can also be used in the arsenal of techniques yielding target-ID clues.