基于子带处理与相关系数的风电机组功率异常波动检测  

Abnormal Power Fluctuation Detection of Wind Turbines Based on Subband Processing and Correlation Coefficient

作　　者：陈刚 胡凯凯 陈亚楠 张家友 熊文豪 CHEN Gang;HU Kaikai;CHEN Yanan;ZHANG Jiayou;XIONG Wenhao(CRRC Zhuzhou Institute Co.,Ltd.,Zhuzhou,Hunan 412001,China)

机构地区：[1]中车株洲电力机车研究所有限公司,湖南株洲412001

出　　处：《控制与信息技术》2024年第2期32-39,共8页CONTROL AND INFORMATION TECHNOLOGY

摘　　要：风电机组功率的周期性异常波动容易引起电网的低频振荡,威胁电网安全;而风电机组功率信号是一种非线性、非平稳信号,其异常波动频率存在不确定性,为异常波动特征的提取与检测带来困难。为此,文章提出一种基于子带处理与相关系数的风电机组功率异常波动检测方法。其首先将原始功率信号通过高通滤波器去除趋势项,然后采用子带处理滤波器组进行信号分解,得到一系列子带信号;将去除趋势项后的功率信号与各子带信号分别计算互相关系数,并根据互相关系数实现周期性异常波动所在子带的定位及子带筛选;基于筛选所得子带信号进行信号重构得到高信噪比重构信号;采用峰值检波与自相关分析相结合的检测方法,识别重构信号中特定幅值周期性异常波动发生时段,从而达到风电机组功率信号周期性异常波动检测的目的。现场数据测试结果表明,对于0.06 Hz、0.32 Hz等不同频率的功率周期性异常波动,该方法能够准确识别其所在子带及特征。Periodic abnormal power fluctuations of wind turbines can easily cause low-frequency oscillations of the power grid,threatening the safety of the power grid.Wind turbine power signals are nonlinear and non-stationary signals,and their abnormal fluctuation Frequencies are uncertain,which makes it difficult to extract and detect abnormal power fluctuation features.This paper proposes a method for detecting abnormal power fluctuations of wind turbine based on subband processing and correlation coefficient.For the application of this method,the author first removed trend items of the original power signal by highpass filter,and decomposed the signal by the subband processing filter bank to obtain a series of subband signals;calculated the cross-correlation coefficient between the power signal with trend items removed and each subband signal respectively,located the subband where periodic abnormal fluctuation occurred and filtered useful subband according to cross-correlation coefficients;obtained high signal-to-noise ratio signal reconstructed based on the filtered subband signal;and by combining the peak detection method and the autocorrelation analysis method,identified the time period where periodic abnormal fluctuations of specific amplitude of the reconstructed signal,thus realizing the detection of periodic abnormal power fluctuations of wind turbine.Field data tests show that for periodic abnormal power fluctuations of different frequencies such as 0.06Hz and 0.32Hz,the proposed method can effectively identify the subband where periodic abnormal fluctuations occur and the periodic abnormal fluctuation characteristics.

关 键 词：风电机组 子带滤波 相关系数 有功功率 周期性波动 

分 类 号：TK83[动力工程及工程热物理—流体机械及工程]