电网阻抗评估翻译.doc

POWER QUALITY RESPONSIBILITIES BY GRID IMPEDANCE ASSESSMENT AT A WIND POWER PRODUCTIONSUMMARY The purpose of this paper is to discuss the grid impedance assessment method at wind farm terminals. The non intrusive method uses variations of the wind power production, and this permanent variability allows us to apply the source impedance identification process, assuring its assessment. New algorithms are developed taking as a reference the accurate GPS synchronized measurement of phase at a very stable power plant, to evaluate the source impedance at a wind farm point of common coupling (PCC) where complex variations of the voltage are expected. The instantaneous value of the grid impedance will, not only, allow assigning responsibilities but also will facilitate the design of equipments, network components, control systems tuning and protections configuration. The development of dispersed generation will be positively influenced by the capacity of measuring the network strength. 在风力发电量上通过电网阻抗评估电能质量 概要 本文的目的是探讨在风力发电厂的电网阻抗的评估方法。非侵入性的方法使用于各种不同的风力发电量，这种永久的变化,使得我们可以运用源阻抗识别过程,保证其评估。在一个稳定的电厂新算法被发展作为精确GPS同步测量阶段，评价在风力发电场源阻抗连接点点(PCC)的预期的复杂变化的电压。电网阻抗的瞬时值，不仅是允许指定的责任而且也要便于设备的设计，网络组件，控制系统的调整和保护的配置。分散一代的发展肯定受网络力量的测量能力的影响。THE TRUE HARMONIC EMISSION LEVEL Standards are ready providing recommendation for preparing the measurements and assessment of power quality characteristics of wind turbines. One important question is raised: Are these fluctuations caused by the wind turbine or by other fluctuating loads? In other words, who is responsible? Generally, the influence of a wind turbine on the voltage quality of the grid depends not only on the turbine but also on the grid where it is connected. The European Wind Turbine Testing Procedure (EWTTP)report has illustrated that the grid impedance angle and short circuit power have an important influence on power quality. We are then facing a very complex problem: how to assess the emission level of a particular distorted generation taking into account the instantaneous fundamental and harmonic source impedance variations of a power network. 真正的谐波辐射水平风力汽轮机发电质量的测量和评估标准已经被准备提供并推荐。一个重要的问题是：风涡轮或者其他波动的负荷会造成波动吗？换句话说,谁负责? 一般来说,风力机的影响,对电网电压质量不仅取决于涡轮还在电网是否连接。欧洲风力发电机测试程序(EWTTP)。报告显示了电网阻抗的角度和电流短路对电能质量有重要影响。我们要面对一个非常复杂的问题:如何评估一个电网瞬时基波和谐波源阻抗变化产生的扭曲的排放水平。Proposed IEC-61400-21 procedureProposed measurement procedures are valid to test the power quality characteristic parameters for the full operational range of wind turbine, connected to a MV or HV network with fixed frequency within 1Hz, sufficient active and reactive power regulations capabilities and sufficient load to absorb the wind power production. The measurements procedures are designed to be as non-site-specific as possible, so that power quality characteristics measured at for example a test site can be considered valid also at other sites. The standard specifies a method that used current im(t) and voltage-time series um(t) measured at the wind turbine terminals distributed over the wind speed interval from cut-in to 15 m/s, to simulate the voltage fluctuations on a “fictitious” grid with no source of voltage fluctuations other than the wind turbine. This fictitious grid is presented in the next figure.提出了IEC-61400-21程序提出了测量过程是有效的测试风力发电设备充分的操作范围内的电力品质特性参数、 在1Hz固定频率内连接到低压还是高压电网，足够的用功功率和无功功率负载能力规定风力发电的生产。测量程序被设计来作为非定点越好, 以便一个试验场可以有效地被用于其它试验场去测量电能质量特性。这个标准规定的方法用于当前电流和电压时间系列测定分布在风速间隔从切入到15米/秒的风力终端，模拟了在“虚拟”电网无源电压波动以外的风力涡轮的电压波动。这个虚拟电网之间的关系用下图表示出来。The fictitious grid is represented by an ideal phase to neutral rated voltage source u0(t) with an appropriate short-circuit apparent power Sk,fic, and four grid impedance angles k while the wind turbine is represented by a current generator im(t), which is the measured instantaneous value of the phase current. Thus, the fluctuating voltage that will be input to the voltage flicker algorithm in compliance with IEC-61000-4-15 to generate the flicker emission value Pst,fic is given as: 虚拟电网被描述在一个理想阶段，当前电流im(t)具有适当的明显的短路电流Sk,fic的额定电压源u0(t)和4角度电网阻抗角k的风力发电机，这是测量的电流的瞬时值。因此, 输入的波动的电压将电压变化算法生成符IEC-61000-4-15标准的瞬时变化值Pst,ficThe fundamental voltage has to be determined from measured voltage at the wind turbine terminals. If voltages and currents are sampled with a constant sampling frequency, the fundamental can be found by filtering the measured voltage and determining the zero crossings of the filtered signal, taking into account phase compensation due to filtering lag.这个基本电压必须取决于风力涡轮终端测量的电压。如果电压和电流以不变的采样频率进行采样，考虑到过滤后的相位补偿，过滤后的被测的电压能够确定零交处的滤波信号能够基本被找到。Once this measurement and simulation procedure is done, flicker coefficients c(k,va) are then reported for continuous operation after normalization and weighting processes depending on the network impedance phase angle k and the annual average wind speed va values. Flicker step factors kf(k) and voltage change factors kU(k) are generated together in the case of switching operations. These factors are normally measured and provided by manufacturers, who trend to overestimate them in some occasions as reported in 3.一旦该测试和仿真程序完成，连续运转规范化，称重过程后变化系数c(k,va)取决于网络的阻抗相角k与年平均风速va值。频率变化因素kf(k)和电压变化因素kU(k) 一起产生的开关操作。这些因素通常是由厂商提供的测量，估计他们的趋势在某些场合中上报的3。It is not the aim of this paper to discuss the current methodology in use for coefficients estimation during the measurement procedure proposed by the standard, but to propose an improvement regarding the second phase of power quality assessment once the wind turbine is connected at the PCC. In order to evaluate the power quality expected from a wind turbine when deployed at a specific site, the following equation applies at continuous operation:它不是在本文中讨论当前的方法在使用期间为系数估计提出的测量方法的标准, 而是提出了一旦风力发电机连接到处理器控制板时改进的第二阶段的电能质量的评估。为了评估来自一个风力发电机部署在某一特定地点时的电能质量,下面的公式应用于连续运行。where Sn is the rated apparent power of the wind turbine and Sk is the short-circuit apparent power at the PCC. Other equations are applied to assess emissions from the sum of several wind turbines, and a similar procedure is proposed to estimate power quality during switching operations 1.在发电机的额定视在功率Sn和处理器控制器上的短路视在功率Sk。其他方程被应用于评估几个风力发电机的输出总和，一个相似的程序被用来估计在转换开关操作时的电能质量。Grid angle dependability is considered by selecting the right coefficient from manufacturers data. Generally, the short-circuit apparent power is assessed by calculation. The nature and the accuracy of data to be provided for modelling the network are not always evaluated in the same way by the experts in charge of harmonics studies. Besides, the source impedance is a time-dependent parameter, varying with a changing network topology, as for example the connection or disconnection of a second parallel transmission line or a second transformer at the bus bar. The development of dispersed generation will lead to important short-circuit level variations as well. We propose to consider short-circuit apparent power variations of the real grid, by evaluating both the fundamental and harmonic source impedance value and its angle. This possibility gives rise to new challenges on the voltage fluctuations assessment, such as more accurate grid-dependent on-line estimation of flicker emission levels. The influence of other fluctuating generators or loads that may cause significant voltage variations at the wind turbine terminals could be identified and responsibilities assigned.电网角度可靠性可以考虑从制造商的数据中选择合适的系数。一般来说，短路电流视在功率是经过计算的。谐波研究的专家并不总是以同样的方式进行电网模型的研究以规定数据的本质和准确性。除此之外, 源阻抗是时变参数，随着网络拓扑结构变化，例如连接或断开平行传输线二次变压器或者一个母线传输二次变压器。离散学的发展也将导致短路电流水平的变化。我们打算通过计算源电源阻抗的基波和谐波的值和角度来分析真正电网上短路电流视在功率的变化。这种可能性引起来了关于电压波动的评估的新的挑战，如在线估计更精确的电网输出闪动排放水平。其他发电机或者负载的影响可能造成负责鉴定和责任分配的风力发电机终端明显的电压变化。Source Impedance assessment process and proposed improvementsOur department has been working in the basis of Yangs method as one of the best ways to measure the fundamental and harmonic source impedance. More details about the process and its practical implementation can be found at 4, 5. Considering a Thevenin equivalent network, a measured current change from Im1 to Im2 between instants t1 and t2, with their consequent voltage change from Vm1 to Vm2 will allow us to determine the true source impedance value. The voltage source Esh value between windows is supposed invariable as well as the source impedance value Zsh. The method is described in the next equations with a load approach: (3) (4) (5)源电源阻抗的评估过程和改进我们部门一直从事基于杨的一种最好的研究方法上测量基波和谐波的源阻抗。更多的过程细节和实际执行安装可以在4, 5找到。考虑戴维宁等效网络, 从时间t1到t2测量电流的变化从Im1到Im2,以及他们从Vm1 Vm2电压变化,将使我们能够确定真正源阻抗值。电压源的值Esh同源阻抗的值一样在图形窗口中被认为是不变的。该方法在接下来的方程中被描述并推导。 (3) (4) (5)Considering the next figure, if the true instantaneous (fundamental and harmonic) value is known, then the normalized value ZshIm/Esh represents physically the true harmonic emission level of an equivalent generation or load.考虑到下图，如果真正的瞬时值Zsh(基波和谐波)是已知的，那么被规范的值ZshIm/Esh描述了物理性上等效负载产生的真正谐波辐射水平。The harmonic voltage emission of the network itself, Esh, can be determined easily from the measured voltage if the harmonic source impedance is known.如果谐波源阻抗值是已知的话，电网本身谐波电压辐射值Esh 就能很容易从被测的电压中被确定Some critical points have been encountered during the implementation phase, and solutions are proposed and evaluated to overcome them.在实施过程中一些关键点已经遇到，并分析提出了解决办法,来克服它们。Accurate phasors calculation: Nowadays, the way to achieve the extraction of fundamental frequency is by means of phase lock loop (PLL) equipment. Sampling frequency is calculated with a delay of 400 ms (20 cycles). The phase shift between windows due to this FFT error may be corrected and compensated.But the system frequency variations remain an important source of errors when trying to be detected by the present technology in use and this is translated in bad phasors calculation. Accurate frequency tracking in a software basis algorithm 6 is one of the new improvements to the process. In this way, 5 or 10 phasors are obtained per (200t) ms window, with t the variable window length depending on the detected frequency, what allows more accurate phasors calculation between windows when determining Zsh 精确向量计算：通过采用锁相环路设备以达到所提取的基本频率。被计算的采样频率有400ms的延迟（20周期）。由于傅里叶变换错误引起的图形相位移动能够被纠正和补偿。但当通过目前使用的技术试图检测时，系统频率变化仍然是一个重要的错误来源，这也将导致错误的相位计算。精确的频率跟踪在软件基本算法上是一种新的过程改进。通过这种方法，能够获得每毫秒200t的5或者10向量，在t时间内，可变波形长度取决于检测频率，在决定Zsh 时在波形中允许更多的精确相位计算。Slow current variations: The method was originally implemented with a delay of 400 ms between windows, not “slow” enough to reflect the overall modification of an event (2s). A continuous current monitoring system is proposed. Some work has been done based on a new algorithm for “current tracking” with the objective of detecting these slow variations and setting minimal thresholds between windows, taking into account the elapsed time between them to prevent any variation on source side (Esh and Zsh).缓慢的电流变化:这种方法最早实施用于400ms的延迟，并不能足够慢的去体现事件的全部改变（2s）。一个连续的电流监测系统被提出了。基于一些新的算法，一些工作已经完成，电流追踪器能检测这些缓慢的变化和设置最小零界值，考虑到运行时间内以防任何其他电源方面的变化(Esh and Zsh)。Synchronization between windows: The process implicates another serious synchronization problem, when supposing the two states for Zsh extraction placed at the same referential related to the system frequency at the time considered. Or in other words, the second measurement window shows a phase displacement of the voltage source Esh. If basic hypothesis are maintained, the referential should remain invariable between windows A synchronization in measurements taken at the terminals of a stable power plant (a nuclear unit for example) and at the wind farm terminals by GPS, is studied in this paper. Two GPS receivers are installed at the nuclear power plant, considered as a reference and at the PCC where the wind farm is connected, as shown in the next figure:波形的同步性：另一个重要的同步问题，当假设在考虑的时间内用相同的参考标准并与系统频率相关的两个Zsh。或者换句话说，第二次的测量波形显示一个电压源的相位移动Esh 。如果基本的假设被保持，那么波形之间的参考仍然被保持不变。在稳定的发电厂终端（以核电厂为例）或者通过GPS的风力发电厂的同步测量在这篇文章中被研究。在核电站和风力发电厂的连接处的处理器控制板上安装两个被视为参考的GPS接收器，在下图中被展示:The proposed synchronization algorithm is represented in the next figure. Obtained results of simulated voltage at the stable infinite bus (nuclear plant E) and at the wind farm terminals (V) time series are represented. 该同步算法在接下来的描述中被提及。在稳定的无线大容量母线上（核电厂）获得的结果和被提及的在风力发电终端的时间序列。As it is seen from the figure, Window 1 (W1) at the nuclear power plant is synchronized with Window 1 (Wsyn1) at the wind farm terminals as for Window 2. Frequency is detected on a 10-cycle basis and phasors are calculated on a 2-cycle basis, obtaining 5 voltage phasors per window. Current time series are inputs to the algorithm at the same time and phasors are extracted at the wind farm terminals. Currents are not represented in the figure. Voltage phasors are computed at the frequency of the nuclear plant, Va1 is computed at fE1 and Va2 is computed at fE2 even if the values of fVa1 and fVa2 frequencies are also calculated and known. Frequency at the wind farm terminals is fluctuating around 50Hz during all the simulation period.从图片中可以看出，核电站的窗口1（W1）与风力发电终端的窗口1(Wsyn1)与窗口2同步。以10周期为基准检测到的频率和以2周期为基准计算出的相量，每个窗口中得到5个电压向量。电流时间序列的算法输入同时和风力发电终端提取的向量。在下表中电流不具备代表性。以核电厂的频率计算出的电压向量，即使fVa1和fVa2是已经计算出来的或者是已知的，Va1和 Va2仍要根据fE1和fE2计算。在仿真阶段风力发电终端的频率在50Hz范围内波动。Several tests have been performed correcting measured voltage phase at the wind farm terminals. Using the same notation as in the figure, the value of the source impedance at fundamental frequency is given by:在风力发电终端几个测试已经被实施校正被测的电压向位。在下图中使用相同的符号，源阻抗基本频率由一下公式给出：The method has been tested on a simplified network and final algorithms have been applied to several simulated wind power schemes. Results are presented in this paper. Using the accurate timing signal provided by Global Positioning system (GPS satellite Universal Coordinated Time (UTC) as common time base for measuring instruments located at both sites we can highly promote the accuracy of synchronized measurements. There are two required outputs of these devices and several optional outputs. The first is a precise digital output with 50 ns rise time that occurs once per second transition of UCT. The second is an ASCII message that identifies the year, day, hour, minute and second of the digital output. Manufacturers are flexible in providing other outputs requested for specialized uses. Because of low frequency of the timing signal of GPS, it can not be used as sampling signal directly. Several options can be adopted for communications like telephone lines, fibre-optic cables, satellites, power lines or microwave links. The synchronization will be achieved locally, at the wind farm terminals by matching the time-stamps of measurements obtained from both sites by a posterior signal analysis. These implementation matters as well as problems related from the communication delays are out of the scope of this work.该方法已被测试用于简化电网和最后的算法已用于一些模拟风力发电计划。在这篇文章中结果被阐释。使用由全球定位系统(GPS卫星环球协调时间(UTC时间)提供的精确定时信号和位于两个地点测量仪表的普通时间基本测量，我们可以更高的提高同步测量的精确性。这些装置有两种必要的输出和几个可选择的输出。第一个是在每秒UTC过度上以每50ns上升时间的精确的数字信号输出。第二个是一种ASCII字符识别的信息,一天,一个小时,一分一秒的数字输出。制造商对于提供需要专门用途的产品时是可以变通的。由于低频率的GPS定时信号,这并不能直接作为采样信号。几个选项可用于通信如电话线、光纤电缆、卫星、电线或微波链接。在本地将获得同步信号，由两个地点的信号进行分析后测量得到的在风力发电厂终端相匹配的时间采集量。这些起动安装的事项与通讯延迟有关的问题都不在工作的范围内。WIND TURBINE SIMULATION RESULTSOne wind turbine unit of 2MW rated power belonging to the BONUS family with a fixed battery of 600kVAr for reactive compensation is connected to an equivalent Thevenin source composed by a voltage source 690V and inductive source impedance. The source impedance in per unit is 0.12590 in a 2MVA - 690V basis. Several tests have been conducted allowing a wide variation of different simulation parameters.风力涡轮机仿真结果一个风力涡轮机组额定功率为2MW属于额外的一个固定的电池600kVAr为无功补偿被连接到一个由电压源690V和感应源阻抗等价戴维南源。源阻抗在2MVA - 690V的基础上以每0.12590。几个测试已经被实施从而允许不同的仿真结果有一个广泛的变化。Wind speed variations: Wind speed as an input to the simulations has been modelled either as a ramp increasing from 10m/s with a wind acceleration of 1m/s or as a step change from 10 to 12 m/s, both speeds having a constant value during at least 5s, having in mind an increasing relatively slowly wind or a sudden step change. More complex simulations have been performed taking as an input a wind speed time series generated from the Kaimal spectra 7. This wind series is generated around a mean speed value, and increased by a ramp, as it can be seen in next figure.风速变化：风速作为一个模拟的输入量已经模拟做以风力1m/s的加速度做10m/s的斜坡上升或者以10到12 m/s的逐步改变，两者的速度至少在5s内有恒定值，有一个相对缓慢的增长或者突然的阶段性改变。更复杂的模拟已经履行了以时间序列的风速产生的Kaimal光谱7作为输入。这种系列的风生成的是一个平均速度值，并以斜坡增长，从下图中可以看出。Damping coefficients: Damping coefficients representing frictional torque at the slow speed shaft, gearbox and high speed shaft have been varied. The wind turbine low frequency oscillations due to the blades mechanical interaction at higher wind speeds is strongly damped for higher coefficients, consequent harmonics are not always present at a wind farm measured current since the wind unit damps higher torsional frequencies. This is illustrated in the next figure.阻尼系数：阻尼系数代表了在低转速下的摩擦力矩，变速箱和高速轴不断多样化。风力发电机的低频振荡是由于机械刀片在较高的风速是相互作用强阻尼系数，由于谐波并非总是出现在一个风力发电场被测量电流从风装置会遏制高扭矩频率。这个在下图中被阐明。Zs algorithm stability: One new objective has been posed with regard to the in-window stability of the process since currents and voltages show important fluctuations within one calculation window. Possible solutions have been proposed to allow the Zsh calculation between two measured windows, such as a minimal threshold of phasor variations. This precondition defines the algorithm sensibility and assures the minimum signal stability for a correct Zs detection. Simulated measurements of wind turbine current and voltage in phase A are performed at a sampling frequency of 20kHz at the wind turbine terminals and resampled to an accurate fundamental phasors detection. Amplitude and phase of voltage and current phasors are then calculated, phase corrected and the Zs calculation is performed every two windows testing di