摘要
介绍了便携式综合电能质量监控系统的组成和工作原理,同时详细说明了各个子模块的硬件、软件结构和在系统中的作用。综合电能质量监控系统融入了具有扰动触发功能的电能质量监测仪,电能质量监测仪可以进行有功、无功、THD等稳态分量测量,同时在发生扰动时记录扰动波形。记录的扰动波形通过CAN总线传输到上位机,通过小波多分辨分析和方差分形计算提取扰动的特征矢量,被提取的特征矢量输入到概率神经网络和支持向量机实现训练和测试。采用了时频分析和模式识别的电能质量监控系统能够实现输、配电系统的瞬态扰动的检测,定位和分类功能,同时系统集成了去噪子程序模块,可以实现噪声条件下的电能质量扰动检测。采用IEEE标准1159的电能质量测试数据,系统对扰动的识别率可以达到90%左右。
A novel portable comprehensive power quality (PQ) monitoring system is introduced in this paper. The paper presents the construction and operating principle of the whole system in detail and expounds the function and theory of each composition module. The instrument system integrates PQ monitoring module with triggering capability, CAN fieldbus and PQ disturbance classifier into PQ monitoring system. PQ monitoring module can measure active power, reactive power, total harmonic distortion (THD) and other stationary components, and simultaneously record the disturbance data under triggering condition. The recorded disturbance data are sent to host computer through CAN fieldbus and the feature vectors of the disturbance signals are extracted by multi-resolution analysis (MRA) of wavelet transform (WT) and variance fractal geometry calculation. Then the feature vectors are inputted into probabilistic neural network and support vector machine for classification of the disturbance signals. The de-noising mod- ule embedded in the monitoring system enables monitoring system to work correctly in noisy environment. The system uses the IEEE Std 1159 to monitor power quality test data and the measurement accuracy can reach about 90%.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2008年第8期1725-1730,共6页
Chinese Journal of Scientific Instrument
关键词
电能质量扰动
电能质量监控系统
小波变换
多维分形
概率神经网络
支持向量机
power quality disturbance
power quality monitoring system
wavelet transform
multi-fractal analysis
probabilistic neural network (PNN)
support vector machine (SVM)