基于奇异值分解和TLS-ESPRIT的电能质量扰动信号检测方法  

Detection Method of Power Quality Disturbances Based on Singular Value Decomposition and TLS-ESPRIT

作　　者：黄奂[1] 黄阳[1] 陈邕安 

机构地区：[1]广西大学电气工程学院,南宁530004 [2]广西国有高峰林场,南宁530001

出　　处：《电测与仪表》2014年第1期64-70,共7页Electrical Measurement & Instrumentation

基　　金：国家自然科学基金资助项目(51107143)

摘　　要：为了能有效地改善电能质量,准确地对电能质量扰动信号的发生及结束时间以及其特征参数进行检测是其前提。文章提出将奇异值分解与TLS-ESPRIT算法相结合来对暂态电能质量扰动进行检测。首先利用基于奇异值分解的方法对暂态电能质量信号进行定位,可以得到较为准确的定位效果。然后以定位点为界,向前及向后各取一个周期的采样信号,对得到的信号利用TLS-ESPRIT算法进行计算,可以估算得到信号的幅值、频率等特征参数。利用这些参数可进一步对扰动类型进行判别。仿真实验结果表明,该方法仅需较短的采样信号,对电压暂降、电压暂升、频率偏差、谐波、谐波加暂降、暂态振荡等多种电能质量扰动信号有较高精度的检测效果。To order to improve the power quality effectively and detect the start time, the end time and the character- istic parameters of the power quality disturbances accurately, this paper proposes a Singular Value Decomposition （SVD） and TLS-ESPRIT-based method to detect the power quality disturbances. First, the star and the end position of the transient power quality disturbance can be determined by the method based on SVD. Then, at each side of the start point in the disturbance signal, one cycle signal is taken out to calculate the characteristic parameters by TLS-ES- PRIT. And the disturbances can be identified by those parameters. The result of simulation tests proves that the meth- od , which needs only the data of a shorter length, is effective for the detection of several kinds of signal of power qual- ity disturbances, such as sags, swells, harmonics, frequency deviation, oscillatory transient ,ect.

关 键 词：奇异值分解 TLS—ESPRIT算法 电能质量 

分 类 号：TM711[电气工程—电力系统及自动化]