基于奇异值分解的电力系统谐波状态估计  

Power system harmonic state estimation based on singular value decomposition

作　　者：徐志向[1] 侯世英[1] 周林[1] 吕厚余[1] 

机构地区：[1]重庆大学电气工程学院

出　　处：《电力自动化设备》2006年第11期28-31,共4页Electric Power Automation Equipment

摘　　要：用传统最小二乘法及其改进方法进行谐波状态估计时,大都是对谐波进行非同步测量,然后求解一个大型的超定线性方程组,其估计精度不足、计算量大、状态量测量数目多且费用昂贵。提出一种基于同步相量测量的谐波状态估计,并用复数奇异值分解求解病态线性复变量方程组的方法,可在系统状态非完全可观的情况下进行有效估计,降低了对测量冗余的要求。以IEEE30节点系统为例,采用同步测量方法测量支路的谐波电流和节点的谐波电压,分别用Matlab和基于奇异值分解(SVD)的最小二乘估计程序进行仿真。结果表明,用SVD算法对系统进行谐波状态估计时较为准确。Traditional and advanced LSE（Least Squares Estimation） methods are mostly to measure harmonics asynchronously and then to solve the giant overdetermined linear equations,which has low estimation accuracy,tremendous calculations and measurements,and high cost. A harmonic state estimation method based on synchronous phasor measurements is presented,and ill-conditioned linear equations with complex variables are solved by the CSVD（Complex Singular Value Decomposition）, which makes the estimation still effective when the system is partial observable and reduces the measurement redundancy degree. Taking the IEEE 30-bus system as an example,branch harmonic currents and bus harmonic voltages are synchronously measured with phasor measurement units. Simulations are carried out with Matlab and the SVD-based LSE program respectively. Results show that the harmonic state estimation based on SVD reflects the system harmonic state more accurately.

关 键 词：谐波 最小二乘状态估计 同步相量测量单元 奇异值分解 

分 类 号：TM711[电气工程—电力系统及自动化]