关于电网供电系统谐波信号检测仿真  

Simulation of Harmonic Signal Detection for Power Grid Supply System

作　　者：曹磊[1] 赵庆生[1] 王旭平[1] 郭尊[2] CAO Lei1 ,ZHAO Qing -sheng1, WANG Xu -ping1, GUO Zun2(1. Shanxi Key Laboratory of Power System Operation and Control, Taiyuan University of Technology, Taiyuan Shanxi 030024, China; 2. School of Electrical and Electronic Engineering, North China Electric Power University, Beijing 102206, China)

机构地区：[1]太原理工大学电力系统运行与控制山西省重点实验室,山西太原030024 [2]华北电力大学电气与电子工程学院,北京102206

出　　处：《计算机仿真》2018年第10期127-131,共5页Computer Simulation

基　　金：国家自然科学青年基金项目(51505317);山西省自然科学基金项目(201601D102039)

摘　　要：谐波检测是实现供电系统谐波治理的前提条件,可提高电能质量。大多数谐波检测算法会由于系统的动态特性导致测量精度的下降。对此,提出了一种基于改进总体最小二乘-旋转矢量不变技术(TLS-ESPRIT)的谐波检测算法。首先根据信号自相关矩阵特征值的相对误差大小得到信号源个数,从而划分出信号子空间。然后用二阶泰勒多项式来表示动态的基波相量,对TLS算法中的范德蒙矩阵进行了扩展;相比TLS-ESPRIT算法,还可估算出基波频率偏移值和频率变化率值。最后采用不同的信号模型和实测数据来检验算法的性能。MATLAB仿真结果表明,所提基于改进TLS-ESPRIT算法能够快速可靠的得到谐波信息。Harmonic detection is the precondition to realize harmonic control in power supply system, and can improve power quality. Most harmonic detection algorithms lead to a decrease in measurement accuracy due to the dynamic characteristics of the system. For this reason, a harmonic detection algorithm based on improved total least squares -estimation of signal parameters via rotational invariance techniques （TLS -ESPRIT） is proposed. Firstly, the number of signal sources was calculated according to the relative error of the signal autocorrelation matrix eigenvalue and the signal space was separated. Then, the two order Taylor polynomial was used to represent the dynamic fundamental phasor, and the Vandermonde matrix of the TLS algorithm was extended, Compared with the TLS - ES- PRIT algorithm, the fundamental frequency offset and the frequency change rate can also be estimated. Finally, different signal models and actual data were used to test the performance of the algorithm. MATLAB simulation results show that the algorithm based on improved TLS - ESPRIT can calculate the harmonic information quickly and reliably.

关 键 词：总体最小二乘-旋转矢量不变技术 谐波检测 动态相量估计 特征值分解 

分 类 号：TM712[电气工程—电力系统及自动化]