基于EMD和PCA的暂态电能质量消噪

Transient Power Quality De-noising Based on EMD and PCA

针对电能质量扰动的消噪问题，提出一种基于经验模态分解（EMD）和主成分分析（PCA）的消噪方法．方法先用EMD将信号分解为一组内蕴模态函数（IMF），对第一层IMF进行细节信息提取，然后对第二层及其后面的IMF进行PCA变换，根据噪声能量选择合适的主成分分量重构，去除各层IMF中的噪声．分别用电压聚降、电压中断、暂态脉冲、谐波及其组合进行数字仿真，和SureShrink小波阈值法、BayesShrink小波阈值法去噪结果比较，所用的方法去噪效果优于SureShrink小波阈值法、BayesShrink小波阈值法去噪结果，尤其对于电压暂降、电压中断、电压聚升这几个最重要的暂态电能质量问题消噪效果更为明显，结果证实了其有效性．

According to the power quality disturbance, a novel de-noising approach based on EMD and PCA is proposed in this paper. At first signal is decomposed as a series of intrinsic mode functions（IMFs） by EMD, the details of the first layer of IMF information are extracted. The PCA transform is performed on the second layer and the back of the IMF, part of principlecomponents are selected to reconstruct the IMF according to the noise energy, then the noise ofeach layer in IMF is removed. Digital simulation is used on voltage sag, voltage Interruption, transient impulse, harmonic and their combined disturbances, the method proposed in this paperis better than SureShrink wavelet and BayesShrink wavelet de-noising algorithm Especially for the voltage sag, voltage interruption, voltage swell which are the most important transient power quality ,the denoising effect is more obvious, and the results confirm its effectiveness.