面向电-热综合能源系统的双线性抗差状态估计方法

提出一种面向电-热综合能源系统(IEHS)的双线性抗差状态估计方法。首先,通过引入辅助变量构建IEHS的线性加权最小绝对值(WLAV)模型,然后通过一次非线性变换与求解一个二次规划模型得到状态变量的估计值。所提方法的优势为：对强相关的多不良数据具有较好的辨识能力;不需要进行非线性迭代,因而具有较高的计算效率。在搭建的IEHS测试系统上验证了所提方法的有效性。     

Tie-line constrained distributed state estimation

This paper presents an implementable distributed state estimation method for online analysis of power systems having multiple, geographically separated areas. Distributed state estimation apart from giving a faster solution, also improves the condition number of the resultant gain matrices. The method proposed here uses the conventional WLS estimator (Gauss-Newton method) with equality constraints forcing the tie line flows as calculated by the adjacent areas to be equal. Based on the topology of the network, the system is partitioned into multiple areas and a processor is assigned to each of these areas for solving the local state estimation problem. Simulations carried out on the IEEE 14, 30 and 118 bus systems show good convergence properties and improvement on the condition number of the gain matrices when compared with centralized algorithms.     

电力系统抗差状态估计研究综述

状态估计是电力系统能量管理系统的重要组成部分.传统加权最小二乘估计不具备抗差性,为此产生了抗差估计理论.本文介绍了抗差估计的定义、目标及基本类型,综述了电力系统抗差状态估计的经典方法和新方法.经典方法主要包括M估计、GM估计、高崩溃污染率估计等;新方法包括新息图状态估计、最大相关熵估计、最小信息损失状态估计、最多赞成状...     

A meter placement method for state estimation

In this paper a new meter placement method is presented. The method is comprehensive in that all the important aspects of the problem-cost, accuracy, reliability, and bad data processing capability requirements-are incorporated into the method. The problem is solved in three stages. The objective in the first stage is to place a minimum set of measurements which satisfies the desired accuracy requirements. In the second stage, additional meters are placed to incorporate reliability requirements which are defined in term of loss of RTUs. In the third stage, additional meters are placed to improve the bad data processing ability of the state estimation in terms of detecting and identifying the bad data     

Least absolute value state estimation with equality and inequalityconstraints

A least absolute value (LAV) state estimator which can handle both equality and inequality constraints on measurements, is developed. It is shown that the use of equality constraints will actually reduce the number of simplex iterations and thus the overall CPU time. The constraints can be used to enhance the reliability of the state estimator without affecting its computational efficiency. The estimation program developed is tested using 14-through 1000-bus power systems     

基于稳健状态估计的设备状态预警研究

针对发电机组设备的状态监测问题,采用改进的自联想回归（AAKR）算法建立设备的状态预警模型。将稳健距离算子代替马氏距离算子,提高自回归模型的稳健性和抗污染能力。将聚类思想应用于距离划分,提出一种基于聚类的变间隔状态矩阵提取方法,并给出具体的实施步骤。采用四重交叉验证学习机制,在训练过程中对模型参数进行优化,获得最优的状态监测模型。以某600 MW机组一次风机为例建立状态参数估计模型,DCS实际运行数据和仿真计算结果表明,改进后的稳健状态估计方法在维持较高准确性的前提下,能够大幅提高模型的稳健性,具有较强的抗污染能力。这样在设备故障早期就能给出相关预警信号,为设备的状态检修提供理论依据。     

用于配电网分布式抗差状态估计的自动分区

网络分区是进行配电网分布式状态估计的基础。分析配电网的网络结构特点,从各分区计算量均衡的角度出发,借助图论中网络的快速搜索能力,通过编程对配电网进行自动分区,从而实现状态估计的并行计算。为改善状态估计算法的抗差性,采用指数表达的权重系数,在迭代过程中降低不良数据在目标函数中权重值,从而减少不良数据对目标函数和估计结果的影响。这种基于分区的配电网分布式抗差状态估计算法可以实现对配电网的自动合理分区,有效地提高状态估计的计算速度,改善了算法对量测量的抗差性。通过IEEE-118节点算例的计算分析,说明了算法的可行性。     

Optimal placement of PMU and SCADA measurements for security constrained state estimation

This paper presents a method for the use of Supervisory Control and Data Acquisition (SCADA) and synchronized measurements for complete observability of a power system. Under normal operation, both Node Phasor Measurement Unit (NPMU) and SCADA measurements are optimally placed using integer programming and Genetic Algorithm (GA) respectively. The minimum condition number of the Jacobian matrix is used as a criteria in conjunction with GA to obtain a completely determined condition. Next, a triangular factorization approach is used to search for the necessary candidates for single branch outage and single/multiple measurement loss. These candidate measurements are optimized by the binary integer programming method. Numerical results on the IEEE test systems are demonstrated. The results clearly show the robustness of the method to obtain reliable measurements under both normal and contingency conditions.     

基于改进Hampel抗差法的电网谐波状态估计

针对电网谐波状态的估计问题,分析了量测粗差对于估计结果的影响,建立了考虑粗差辨识的谐波状态估计数学模型,应用Hampel抗差法对量测数据进行了辨识分类,并根据谐波状态估计的特点对算法的阈值函数进行了改进,有效地解决了"残差污染"和"残差淹没"问题。仿真计算及分析结果表明,求解谐波状态估计问题时,改进Hampel抗差法能有效抑制粗差对估计结果的影响,提高了估计精度。     

基于抗差理论的P-Q分解状态估计算法

状态估计是能量管理系统(EMS)的重要组成部分,提高状态估计的计算精度具有重要意义,基于此,在当前状态估计计算中广泛应用的P-Q分解法的基础上,应用抗差估计理论中的极大似然型估计(M估计),构造出一组权函数,从第二次迭代计算开始,根据每次迭代计算后各量测量残差与量测量自身之比的大小,对相应加权值进行修正,使不良数据的权值越来越小,达到弱化不良数据对后续迭代过程影响的目的。最后,以某地区的220 kV电网为实际背景,通过Matlab编程仿真,与P-Q分解的估计数据相比较,证明该改进方法简单实用,并在提高状态估计的估计精度和自动弱化不良数据影响上有很好的效果。     

基于抗差理论的P-Q分解状态估计算法

状态估计是能量管理系统(EMS)的重要组成部分,提高状态估计的计算精度具有重要意义,基于此,在当前状态估计计算中广泛应用的P-Q分解法的基础上,应用抗差估计理论中的极大似然型估计(M估计),构造出一组权函数,从第二次迭代计算开始,根据每次迭代计算后各量测量残差与量测量自身之比的大小,对相应加权值进行修正,使不良数据的权值越来越小,达到弱化不良数据对后续迭代过程影响的目的.最后,以某地区的220 kV电网为实际背景,通过Matlab编程仿真,与P-Q分解的估计数据相比较,证明该改进方法简单实用,并在提高状态估计的估计精度和自动弱化不良数据影响上有很好的效果.     

Application of a robust algorithm for dynamic state estimation of apower system

In this paper, a computation algorithm using the exponential function to increase the robustness of the dynamic state estimation is proposed. The merit of this approach lies in its immunity to the polluted measurements, while the implementation of the method is not complicated when compared with other methods. To validate the effectiveness of the proposed method, it was tested through several example power systems under different scenarios that include normal operation, bad measurements, sudden load change, and topology error conditions. From test results, they help support the feasibility of the method for state estimation applications     

一种适应于配电网的交叉逼近状态估计方法

配电网的高阻抗比特点使得传统的输电网快速解耦状态估计无法使用,且快速解耦状态估计无法处理配电网中大量存在的电流幅值量测,因而往往不满足可观性要求。为此,提出一种适应于配电网的交叉逼近状态估计方法,通过交叉迭代求解所有节点电压幅值的平方和所有支路两端的相角差,待收敛后再求解电压复相量。算例分析表明所提方法对高阻抗比的配电网具有很好的适应性,在满足计算精度要求的同时具有较高的计算效率。     

电力系统多区域分布式状态估计方法

提出了一种电力系统多区域分布式状态估计方法,各区域估计器利用其数据采集与监视控制系统提供的量测数据进行本地状态估计,并通过平均一致性算法获取全局信息进行系统级状态估计。建立了基于拉格朗日乘子法的状态估计模型并设计了基于一致性的全局信息交换协议,给出了多区域分布式状态估计算法的实现流程。通过IEEE 14节点和118节点系统中的仿真算例验证了所提方法的正确性和有效性,并就估计精度和计算效率与现有状态估计方法进行了比较。仿真结果表明分布式状态估计方法可有效提高集中式状态估计系统的计算效率及可靠性,适用于结构更加复杂、量测数据体量更大电网的状态估计。     

A constrained orthogonal state estimator for external system modeling

Attention is focussed on the application of constrained orthogonal state estimation to external system modeling. Inequality constraints enforce realistic limits on the estimates of External System (ES) variables that otherwise can become off bounds due to the limited ES monitoring capabilities at the Internal System control center. Equality constraints are used to model known zero bus injections in both systems. The use of orthogonal methods is advocated to ensure numerical stability, since the estimator has to handle a larger range of measurement weights than in conventional applications of state estimation. Tests to evaluate the performance of the proposed constrained estimator using three power networks of different sizes are also reported in the paper.     

基于抗差总体最小二乘法的电力系统谐波状态估计

介绍了谐波状态估计的数学模型及最小二乘和总体最小二乘的求解算法,综合考虑测量误差和参数误差的影响,利用总体最小二乘法进行谐波状态估计。分析了测量粗差对估计结果的影响,针对总体最小二乘法不具备抑制粗差能力的缺点,提出利用抗差总体最小二乘法进行谐波状态估计。用Matlab搭建了IEEE-14节点谐波测试系统仿真模型,在测量数据和参数矩阵中分别加入含有粗差的正态分布误差及正态分布误差,画出概率密度曲线图,并对总体最小二乘法、抗差最小二乘法和抗差总体最小二乘法进行比较,结果表明利用抗差总体最小二乘法能够得到更精确的谐波状态估计结果。     

新息图法状态估计

提出了基于新息图的状态估计方法,可以快速地检测识别出拓扑错误,进行状态估计,提供电力系统状态.     

新息图法状态估计

提出了基于新息图的状态估计方法 ,可以快速地检测识别出拓扑错误 ,进行状态估计 ,提供电力系统状态     

基于分块雅可比矩阵的最小二乘状态估计算法

提出了一种基于分块雅可比矩阵的加权最小二乘状态估计算法.该方法将全部的注入功率量测、赋予很大权值的虚拟量测和必要的支路量测来构造一组恰好可求解系统全部状态变量的量测集,将余下的支路功率量测作为量测系统的冗余部分看待,并依此对量测雅可比矩阵进行分块,在此基础上推导出了该文算法迭代的修正方程式.该文所提算法既具有很好的数值稳定性,又能够减少计算量.仿真试验表明该方法是快速且数值稳定.