【作者】 杨静；

【导师】 吴敏； 陈通文；

【作者基本信息】 中南大学 ， 控制科学与工程， 2012， 博士

【摘要】 电力系统的发展和大区联网的实现,使得电力系统规模日益扩大、电网结构更加复杂,为国民经济带来显著效益的同时,也使电力系统运行点越来越接近稳定极限,整个系统发生电压失稳甚至电压崩溃的危险不断增加。电力系统仿真研究和各种具体分析都是建立在相应的数学模型基础上,各元件精确的数学模型能更准确地反映实际系统的动态过程,提高计算精度与可信度,有助于保证系统稳定安全运行、并提高分析决策水平,在电力系统的运行、控制、计算中具有现实意义。为了实现电力系统稳定安全运行,本文重点研究电力系统传输网络模型和非线性动态负荷模型的辨识及其在电压稳定分析中的应用。论文取得的主要研究成果和创新点如下：1)提出改进的带遗忘因子的最小二乘法及其病态问题改善方法,实现多源、多负载电网模型的在线辨识。针对复杂电力系统传输网络,基于基尔霍夫电压、电流定律,引入电压增益矩阵和电网阻抗矩阵,建立顾及线路耦合关系的多源、多负载电网稳态线性模型。与戴维南等值模型相比,该模型更利于理解和研究系统的整体运行效果。为了适应电网参数经常变化的特点,基于同步相量测量数据,采用带遗忘因子的递推最小二乘方法辨识该电网模型,实现电网阻抗参数的估计和跟踪；同时提出一种改进的基于奇异值分解的病态问题改善方法,提高了模型的辨识精度和可信度。最后通过两个简单电力系统进行仿真实验,结果表明本文所提方法可以有效用于多源、多负载电力系统传输网络模型的在线辨识。2)提出基于电网模型的线路故障诊断和电压稳定分析方法基于多源、多负载电力系统传输网络模型,研究其在线路故障诊断和电压稳定分析等方面的应用。首先根据电网模型计算并跟踪系统导纳矩阵的变化,根据矩阵各元素的变化情况,提出线路故障诊断方法。其次对耦合的电网模型进行解耦处理,提出一种静态电压稳定指标,表征系统当前运行点与稳定运行极限点的距离,方便判定各负荷节点的电压稳定性及薄弱程度,利于整个电网稳定监测。最后采用简单5节点系统和IEEE-30节点标准电力系统对所提出的线路故障诊断和电压稳定性分析方法进行验证。结果表明本文所提方法可以有效的判断系统线路故障发生点,并实时监测系统稳定运行状态。3)提出电力系统非线性动态负荷模型辨识方法针对Hill等提出的指数功率恢复模型的非线性和动态特性,采用二阶泰勒级数展开模型作为该模型的简化模型,提出一种基于频率加权最小二乘的Hartley调制函数算法进行模型参数的辨识。这是一种针对连续时间系统的直接辨识方法,该方法巧妙的避免了信号的数值微分处理及初始条件的确定,克服了传统离散系统辨识方法在连续系统模型和离散系统模型相互转换时引入的误差以及对采样周期的依赖。选择合适的调制函数参数可以有效提高系统辨识的抗噪能力。为了克服Hartley调制函数方法参数选择复杂的缺点,简化辨识过程,充分考虑指数功率恢复模型的结构特征,本文提出一种基于离散傅里叶变换数值微分的交替最小二乘辨识方法。将非线性负荷模型辨识问题分解为两个线性最小二乘问题,采用交替辨识方法逐步估计负荷模型参数。针对负荷模型辨识过程中需要计算数值微分的特点,提出基于离散傅里叶变换的数值微分方法,有效减小数据噪声对辨识结果的影响,而且提高了计算速度。仿真结果表明,本文所提方法在含有明显噪声情况下仍然可以获得比较精确的辨识结果。4)提出考虑非线性动态负荷模型动、静态特性的电压稳定分析方法结合P-V曲线法,分析动态负荷模型参数对系统运行点位置的影响,研究负荷静态特性和动态特性如何影响着电压失稳事故的发生,提高了系统的稳定运行监控能力。通过实例分析,证明了系统在相同条件下,负荷特性将会影响P-V曲线上系统运行点的位置,从而导致系统离电压崩溃点更近或者更远。通过实时监测系统运行状态,在系统接近稳定运行极限时,及时采取无功补偿或切负荷等措施,防止发生电压失稳和电压崩溃等事故的发生。更多还原

【Abstract】 The past decade has seen major restructuring and deregulation in the electric utility industry; significant changes have been taking place, bringing in the notion of competition and market strategies in the operation of electric power systems, and the need to reduce operating and capital costs. This economic pressure drives the trend to maximize the utilization of high-voltage equipment, operating at conditions closer to the system limits; however, this also put systems at a higher risk for collapse. For safe and efficient operating of large power systems, it is of paramount importance to provide real-time and accurate estimates of the voltage stability margin. Accurate grid models and load models are important in monitoring and protection of wide-area power systems.This dissertation addresses issues in the estimation and application of power grid impedance matrices and dynamic load models based on synchronized phasor measurements, for improving monitoring and better operation of large power systems. The main results and innovations are listed as follows:1) Online impedance estimation of power grids using a recursive least squares estimation algorithm with a forgetting factor.For better understanding overall operation and performace of entire power systems, a multi-bus power grid model is obtained based on Kirchhoff’s current and voltage laws. A recursive least squares estimation algorithm with a forgetting factor is presented to identify impedance parameters of the model. This approach is based on synchronized phasor measurements and makes it possible to track changing parameters. The algorithm is tested on two power systems using MATPOWER-generated data, and case studies are conducted for estimation of grid impedances. The results show that the proposed method is effective for online estimation of grid impedances.2) Applications of grid impedance matrices to voltage stability margin analysis and fault detection.Application of power grid impedance matrices to voltage stability margin analysis and fault detection is studied. Based on the Kirchhoff’s law, the system admittance matrix can be obtained from the grid model. Then a method to dectect transmission line outage by observing grid impedance changes is proposed. A multi-port method for voltage stability margin prediction is proposed based on the estimated network impedance matrix. A simple5-bus system and the IEEE30-bus power system are used to investigate the proposed methods, and the results showed that the proposed method can be utilized for dectecting transmission line outage and online monitoring of voltage stability with high accuracy.3) Identification of nonlinear dynamic load models of power systems.A second order Taylor series expansion model is presented as the simplification of the popularly used exponential recovery dynamic load model. Then, a frequency weighted least-squares based Hartley modulating functions method is developed to estimate parameters of the load model, which effectively attenuates the effect of noises in the process of parameter identification. However, it is hard to choose the optimal parameters of the Hartley modulating fuction method.To simplify the identification process and to improve the identification accuracy, considering the special characteristic of the exponential recovery dynamic load model, an alternating least squares approach is developed to decompose the nonlinear optimization problem of dynamic load parameter identification to two linear least squares problems, which can be solved in an alternating manner. Then, a DFT-based numerical derivative technique is proposed to attenuate the effect of noises in the process of parameter estimation. Simulation studies are conducted to demonstrate the accuracy of the proposed methods for identifying the dynamic load model in the presence of noticeable additive measurement noise.4) Application of the dynamic load models to stability margin analysis.P-V curves play an important role in understanding voltage instability. Under the same conditions in a power system, load characteristics affect the location of operating point in the P-V curves, which is distinctly important for voltage stability analysis. The influence of dynamic load characteristics on voltage stability is studied. Simulation analysis showed that load characteristics are critical for the stability of the system. The parameters of the dynamic load model are helpful for voltage stability analysis and protection.更多还原