基于自适应阈值和奇异值分解的电能质量扰动检测新方法  

A New Detection Approach of Power Quality Disturbances Based on Adaptive Threshold and Singular Value Decomposition

作　　者：杨晓梅[1] 罗月婉 肖先勇[1] 郭朝云 YANG Xiaomei;LUO Yuewan;XIAO Xianyong;GUO Chaoyun(School of Electrical Engineering and Information, Sichuan University, Chengdu 610065, Sichuan Province, China)

机构地区：[1]四川大学电气信息学院

出　　处：《电网技术》2018年第7期2286-2294,共9页Power System Technology

摘　　要：随着电力系统规模的增大,为保证电力设备的稳定运行,电能质量数据的分析算法面临更高的要求。在考虑算法精确度及抗噪性的同时,对一些微弱扰动及复杂扰动的检测也提出了新的挑战。针对以上需求,提出一种通用的电能质量扰动检测算法,基于异常扰动波形的特点,提出基于波形差值的自适应阈值,无需调节参数即可对信号进行故障判断,实用性高,并且采用加窗的方式对信号进行奇异值分解,并根据检测结果给出明确的扰动定位;最后通过多组模拟信号与实测信号的仿真实验,验证了所提算法的有效性,并证明了算法具有计算量小、实时性高的优点;通过与其他算法的对比分析,进一步表明了文中所提算法具有更高的灵敏度与抗噪性,普适性强。With development of power system capacity, analysis algorithms of power quality data faces higher requirement to ensure stable operation of electric equipment. While considering accuracy and noise resistance of algorithms, new challenges are also raised for detection of some weak or complex disturbances. In view of above requirements, in this paper, a general detection algorithm of power quality disturbances is proposed based on characteristics of abnormal disturbance waveform. The proposed algorithm combines sliding window and singular value decomposition, and a definite disturbance location can be given according to detection result. Meanwhile, an adaptive threshold based on waveform difference is also proposed for distinguishing abnormality from noise without need of adjusting parameters. Finally, effectiveness of the proposed algorithm is verified with simulation for a number of simulated and actual signals, showing that the algorithm has advantages of small computation amount and high real-time performance. Comparing with other algorithms, the proposed algorithm has higher sensitivity and noise resistance capability, applicable to various situations.

关 键 词：电能质量 异常扰动 波形特征 奇异值分解 自适应阈值 

分 类 号：TM721[电气工程—电力系统及自动化]