一种电能质量扰动信号的联合去噪算法  

A Combined De-Noising Method for Power Quality Disturbances Events

作　　者：肖贤贵 李开成[1] 蔡得龙 王梦昊 王伟[1] Xiao Xiangui;Li Kaicheng;Cai Delong;Wang Menghao;Wang Wei(State Key Laboratory of Advanced Electromagnetic Engineering and Technology Huazhong University of Science and Technology,Wuhan 430074 China;State Key Laboratory of Power Grid Environmental Protection Wuhan Branch of China Electric Power Research Institute Co.Ltd,Wuhan 430074 China)

机构地区：[1]强电磁工程与新技术国家重点实验室(华中科技大学),武汉430074 [2]中国电力科学院有限公司武汉分院电网环境保护国家重点实验室,武汉430074

出　　处：《电工技术学报》2021年第21期4418-4428,共11页Transactions of China Electrotechnical Society

基　　金：国家自然科学基金资助项目(52077089)。

摘　　要：传统的小波阈值去噪算法存在易丢弃真实信号、去噪效果差等缺点。该文提出了一种新的用于电能质量扰动事件的联合去噪算法。该算法首先通过强跟踪卡尔曼滤波的渐消因子大于1的次数,初步判定信号的扰动类型,然后对不同的扰动类型采取不同的去噪方法。对于仅含噪声的正弦信号和谐波信号用稀疏分解及快速傅里叶变换(FFT)做两次去噪;对暂升和暂降信号采用渐消因子准确地指示扰动起止时刻,将信号分段,并对每段信号用稀疏分解和FFT去噪;对含瞬态脉冲和暂态振荡的信号,采用不同的处理方法,首先通过稀疏分解得到稳态成分和暂态成分,对稳态成分的去噪方法与含噪声正弦信号的处理方法相同,对暂态成分的脉冲信号保留实际值,对振荡信号采用变分模态分解(VMD)去噪。大量的仿真计算表明,在不同的信噪比条件下,该文提出的算法均能够有效抑制各类扰动信号的噪声,显著提高了信噪比,且效果优于小波阈值去噪算法。The traditional wavelet threshold de-noising algorithm has some disadvantages, such as easy to discard the real signal and poor de-noising effect. This paper presented a new joint denoising algorithm for power quality disturbance events. Firstly, the disturbance type of signal is preliminarily determined by the number of times that the fading factor of strong tracking Kalman filter is greater than 1, and then different de-noising methods are adopted for different disturbance types.For sinusoidal signal and harmonic signal only containing noise, sparse decomposition and fast Fourier transform(FFT) are used to de-noising twice;fading factor is used to accurately indicate the beginning and ending time of disturbance for sag and rise signal, and sparse decomposition and FFT de-noising are used for each segment of signal;different processing methods are used for signal containing transient pulse and transient oscillation. Firstly, the steady-state component and transient component are obtained by sparse decomposition. The de-noising method of steady-state component is the same as that of sinusoidal signal. The impulse signal and oscillation signal of transient component are processed by retaining the actual value and variational mode decomposition(VMD) respectively. A large number of simulation results show that under different SNR conditions, the proposed algorithm can effectively suppress the noise of various disturbance signals, significantly improve the signal-tonoise ratio, and the effect is better than the wavelet threshold de-noising algorithm.

关 键 词：电能质量扰动 强跟踪卡尔曼滤波 稀疏分解 小波分析 去噪方法 

分 类 号：TM71[电气工程—电力系统及自动化]