采用改进互补集总经验模态分解的电能质量扰动检测方法

针对集总经验模态分解（EEMD）方法加噪参数（噪声幅值、集总次数）需人为确定、分解残余噪声大以及计算耗时长的缺点，提出一种自适应快速互补集总经验模态分解（AFCEEMD）方法.该方法分析不同频率形式噪声对极值点分布的影响，确定加噪频率采用高频辅助分解的优势，并以极值点分布特性作为评价指标自适应选择最优加噪频率.通过对EEMD加噪准则的研究，推导出加噪幅值和分解次数采取固定值：0.01 SD和2次，且以正负成对的形式加入到原始信号中.通过仿真实验和搭建的电能质量扰动平台的实测数据验证了所提方法的自适应性和计算性能，而且适用于电能质量扰动检测与分析.
Abstract: An adaptively fast complementary EEMD (AFCEEMD) was proposed aiming at the shortcomings of the ensemble empirical mode decomposition (EEMD) method, where the two critical parameters (the amplitude of the added white noise and the number of ensemble trials) were obtained artificially, the contamination of residue noise in the signal decomposition and the computational complexity. In the proposed method, the influence of white noise with different frequency forms on the distribution uniformity of signal extreme points was analyzed, the advantage of employing high-frequency white noise auxiliary decomposition was determined, and the distribution characteristic of signal extreme points was taken as an evaluation index to adaptively select the optimal frequency of additive white-noise. Furthermore, the two key parameters respectively fixed as 0.01 times standard deviation of the original signal and two ensemble trials were deduced by investigating the principle of added white noise in EEMD method, in which the white noise was added in pairs with plus and minus signs to the targeted signal. Both simulation tests and experiments data measured from the built power quality disturbance platform demonstrate the adaptability and computational efficiency of the proposed method, which is suitable for detection and analysis of power quality disturbance.