基于改进HHT的电压闪变检测

针对应用希尔伯特-黄变换(HHT)算法进行电压闪变参数检测过程中经验模态分解(EMD)产生的固有模态分量(IMF)不理想而增大参数检测误差的问题,提出了一种改进HHT的电压闪变检测方法。首先通过在EMD"筛选"步骤中添加四点插值细分算法"分裂"出新的控制点供三次样条插值拟合包络线,然后分解出一组IMF分量,最后对IMF分量采用Hilbert变换得到闪变检测参数。就含噪声的单一分量闪变信号、不含噪声的和含噪声的多分量闪变信号,分别采用未改进HHT方法与改进HHT方法进行仿真检测,研究结果表明,改进HHT算法具有良好的抗噪能力,对模态混叠具有一定的抑制作用,并且能够提高闪变信号参数检测的准确度。

Voltage flicker parameters detection based on modified HHT

During the detection of voltage flicker parameters using HHT algorithm, the intrinsic mode functions ( IMF) generated by empirical mode decomposition ( EMD) are not ideal and the parameter detection errors are grea-ter.This paper presents a modified HHT detection method of voltage flicker.Firstly, by adding the four point interpo-lation subdivision algorithm in the EMD "sifting" steps, a group of control points are split and the fitting curve is formed by cubic-spline interpolation, and then, a set of IMF components are decomposed.Finally, the Hilbert trans-form is used to obtain the voltage flicker parameters.As to single component flicker signal containing noise and multi-component flicker signal containing noise or no noise, the simulation experiment was carried out by using HHT method and modified HHT method respectively.The research results show that the modified HHT algorithm has better anti-noise ability, which can suppress the modal aliasing and improve the accuracy of flicker signal parameters detection.