电力系统输电断面及故障评估关键技术研究

发布时间：2018-03-30 00:37

本文选题：电力系统　切入点：输电断面　出处：《华中科技大学》2014年博士论文

【摘要】：保障电网安全高效运行是现代电力系统发展的主要目标之一。近年来,高性能计算机技术已能实现上万节点规模的大电网分钟级数据分析,监控与数据采集系统／能量管理系统(SCAD A/EMS)和广域量测系统(WAMS)可为电网控制中心提供多时间尺度的信息源,这些都为跨大区电网的在线评估提供了可能。本文以多信息源为基础,以电力系统输电断面及故障评估关键技术为研究目标,取得的主要成果如下：在研究了现有输电断面研究在应用层面上的局限性的基础上,提出了两种新的输电断面快速识别方法。通过分析支路开断下的潮流转移特性,结合图论割点与块的性质,在总结归纳潮流转移断面图论特征的基础上,提出了基于割点与路径搜索的输电断面识别方法。此外,利用电气距离选取了系统区域联络线,进而通过联络线节点进行系统分区,最终根据区域间的连接割集及潮流一致性特征提出了基于输电通道潮流一致特征的输电断面识别方法。通过对这两种输电断面识别方法的研究,建立了输电断面的多角度分析体系。针对系统薄弱断面快速定量评价还存在的若干问题,提出了一种基于电压崩溃指数的极限传输容量计算模型和实用化算法。该方法与传统连续性方法相比具有如下四个特点：以原始牛顿法为潮流计算核心,不会出现连续潮流扩展雅可比矩阵奇异的现象；根据雅可比矩阵信息计算相应的电压崩溃指数,该参数可以引导整个状态推演过程,将系统负荷(或发电)状态准确定位至功率极限点；利用雅可比矩阵信息自动选取合适的状态推演步长,确保全网负荷(或发电)变化能够准确地向系统临界点逼近；整个推演过程以雅可比矩阵为计算核心,不需要每步状态推演都有完整的牛顿迭代过程,从而使计算速度大幅提升。由于传统常规潮流计算方法计算过程耗时较大,难以应用于极限传输容量及薄弱断面的快速评估这一难题,提出了一种基于非迭代潮流计算的极限传输容量计算模型和实用化算法。该方法把Taylor级数引入到电力系统潮流计算当中,它能够实现系统极限传输容量快速、实时求解的目标。传统故障定位算法都基于理想化的测量信息条件,这使得其在实际电网的应用中面临着众多问题。本文提出了基于故障可观性的输电线路故障定位方法,它提供了一种在电网在信息不全条件下的故障定位方案。该方法采用分布参数输电线路模型,可以更准确地考虑分布电容电流对故障定位精度的影响；算法性能优异,且不受故障初始相位角、负荷及故障路径阻抗的影响,仅需要相量测量单元(PMU)的暂态数据信号,无需系统阻抗等其它信息,算法没有迭代过程,计算速度快。考虑到传统信号分析方法在处理电力系统故障暂态信号过程中误差较大的问题,本文提出了一种基于快速经验模式分解(EMD)滤波的自适应相量评估算法,该算法可以自适应地滤除电力系统故障信号中的衰减直流分量。为了满足实时计算的要求,本文在筛选过程、最优包络拟合及数据窗选择等方面对EMD算法做了重要改进。
[Abstract]:Ensure safe and efficient operation of power grids is one of the main objectives of the development of modern power system. In recent years, the high performance computer technology has been able to achieve the scale of tens of thousands of nodes power minute level data analysis, monitoring and data acquisition system and energy management system (SCAD A/EMS) and wide area measurement system (WAMS) can provide multi time scale power grid the control center of the information source, which provides the possibility for online assessment of inter regional power grid. Based on the multi information source, with the key technology of power system transmission section and fault evaluation as the research object, the main achievements are as follows:
In the study of the existing research limitations in the transmission section based on application level on the proposed two new fast recognition method of transmission section. Through the analysis of branch breaking under flow transfer characteristics, combined with the properties of graph cut point and block, based on summing up the characteristics of power flow transfer section graph theory, put forward the recognition method of transmission section cut point and based on the path search. In addition, the use of electrical distance system regional contact line, and then through the contact line system partition node, according to the final cut set and connected trend consistent characteristics between regions proposed recognition method of transmission channel transmission section based on the characteristics of the trend consistent through the research on this. Two kinds of identification methods of transmission section, a transmission section of the multi angle analysis system.
In order to solve the problems of rapid quantitative evaluation system has weak cross section, presents a calculation model and method to limit the transmission capacity of voltage collapse based on the index. Compared with the traditional continuous method has the following four characteristics: the original Newton method for power flow calculation of the core, not a continuous trend of extended Jacobi matrix is singular according to the information phenomenon; Jacobi matrix calculation of voltage collapse index corresponding, the parameter can guide the state inference process, the system load (or power) state of accurate positioning to the power limit point; using the Jacobi matrix information automatically selects the appropriate state inference step, ensure that the entire network load (or power) can accurately approximate the changes to the system critical point; the deduction process using the Jacobi matrix as the computing core, does not need each state has complete deduction of Newton iteration Process, so that the speed of calculation is greatly improved.
The calculation method of conventional power flow calculation process is time-consuming, the problem of rapid assessment is difficult to be applied to limit the transmission capacity and weak cross section, a method is proposed for calculating the non iterative power flow based on the limit of transmission capacity calculation model and practical algorithm. The method of the Taylor series was introduced into the power flow calculation of power system, it can realize the system the limit of transmission capacity fast, real time solution.
The traditional fault location algorithms are based on the measurement information of idealized conditions, which makes it faces many problems in the practical application of power grid is put forward in this paper. Based on the fault of transmission line fault location method is considerable, it provides a grid in the incomplete information under the condition of fault location scheme. This method adopts distributed parameter the transmission line model can be more accurate to consider the influence of the distributed capacitance current on the fault location accuracy; algorithm performance is excellent, and is not affected by the fault initial phase angle, impact load and fault path impedance, only phase measuring unit (PMU) of the transient data signal, without system impedance and other information, no algorithm the iterative process, fast calculation speed.
Considering the traditional signal analysis of the large errors in the processing of fault transient signals in power system process, this paper presents a fast empirical mode decomposition based on adaptive filtering (EMD) phase estimation algorithm, this algorithm can adaptively filter the fault signals in power system the decaying DC component. In order to meet the real-time computing requirements. In the screening process, optimal envelope fitting and data window made important improvements to the EMD algorithm.

【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TM732

【参考文献】