运用因散经验模式分解算法的谐波检测新方法  

A novel method of detecting harmonic currents using EEMD algorithm

作　　者：成立[1] 吴衍[1] 杨宁[1] 王鹏程[1] 王振宇[1] 

机构地区：[1]江苏大学电气与信息工程学院,江苏镇江212013

出　　处：《江苏大学学报（自然科学版）》2010年第6期687-690,715,共5页Journal of Jiangsu University：Natural Science Edition

基　　金：国家"863"计划项目(2006AA10Z258)

摘　　要：为了满足有源电力滤波器(APF)实时、快速地跟踪检测电力系统谐波电流的要求,提出基于因散经验模式分解算法的谐波电流检测法.该方法将电流信号分解成内在模式函数(IMF)形式,在筛分过程中添加高斯白噪声频谱,为IMF解析过程的时域分布设定一致的参考结构,并运用端点效应处理策略确定边界极值点.设计了运用新算法的电流跟踪检测器,并进行了仿真试验.结果表明:该方法不仅在筛分时去除了模式混叠,而且硬件配置简单,检测基波幅值与期望幅值之误差仅为1.08%,可以较精确地分解出电流信号的基波和谐波分量;该方法跟踪检测非平稳信号的延时仅为6μs,信号的谐波分析实时性比传统的谐波检测法优越,因而该法可用于APF的电流跟踪控制电路和其他的谐波电流检测器.To meet the power system′s needs of real-time,fast tracking and detecting of harmonic currents with an active power filter（APF）,a novel method of detecting the power system harmonic currents based on ensemble empirical mode decomposition（EEMD） was proposed.The current signal was decomposed in terms of intrinsic mode functions（IMF）.The Gaussian white noise was adopted in sifting process,in order to provide a uniform reference frame in the time space.End effect process strategy was adopted to solve the mode mixing problem.A current tracking detector was designed based on this algorithm and the simulation was carried out.The results show that the method can eliminate the mode mixing in the sifting process,the hardware is simpler,and the error between the detected and the expected amplitudes of fundamental component is only 1.08%.The algorithm can separate the current signal into fundamental component and harmonic components accurately,and the delay-time of non-stationary signals tracked and detected by the algorithm is only 6 μs.The real-time harmonic analysis of the algorithm is superior to the traditional,thus the method can be applied to current tracking control circuits of APF and other harmonic currents detectors.

关 键 词：电力系统 有源电力滤波器 EEMD算法 谐波电流检测法 内在模式函数 

分 类 号：TM835.4[电气工程—高电压与绝缘技术]