自适应陷波滤波器频率估计新方法及性能分析

发布时间：2018-07-04 13:38

本文选题：自适应陷波滤波器 + 频率估计　；参考：《电子学报》2014年01期

【摘要】：现有的自适应陷波滤波器(ANF)受误差函数所限,导致其自适应频率估计方法收敛速度较慢,对初始迭代频率值设定范围要求较高,特别针对频率接近于0或π的信号,还存在频率估计精度不高、算法稳定性差的问题,为此,提出一种ANF频率估计新方法.首先,分析现有ANF方法估计信号频率时存在精度低、速度慢、稳定性差的原因,提出一种新误差函数以提升ANF收敛速度;然后,根据ANF估计信号频率时偏差产生的机理,通过偏差补偿方式,降低噪声对ANF的影响,以获得近似无偏的频率估计结果,提高ANF频率估计精度,同时与离散卡尔曼滤波相结合,以改善算法的稳定性,并对该方法进行稳态条件下的性能分析;最后,给出了ANF频率估计结果,并讨论了ANF各参数对频率估计精度的影响,给出了具体计算结果.计算表明本文方法的有效性与正确性.
[Abstract]:The existing adaptive notch filter (ANF) is limited by the error function, which leads to the slow convergence speed of its adaptive frequency estimation method, and requires a high range of initial iterative frequency values, especially for signals whose frequencies are close to 0 or 蟺. There are also the problems of low accuracy of frequency estimation and poor stability of the algorithm. Therefore, a new method of ANF frequency estimation is proposed. Firstly, the reasons of low precision, slow speed and poor stability in the estimation of signal frequency by existing ANF methods are analyzed, and a new error function is proposed to improve the convergence speed of ANF, and then, according to the mechanism of deviation in estimating signal frequency by ANF, a new error function is proposed to improve the convergence speed of ANF. The influence of noise on ANF is reduced by offset compensation method, and the approximate unbiased frequency estimation results are obtained, and the accuracy of ANF frequency estimation is improved. At the same time, the stability of the algorithm is improved by combining with discrete Kalman filter. Finally, the results of ANF frequency estimation are given, and the influence of ANF parameters on the accuracy of frequency estimation is discussed. The calculation shows the validity and correctness of this method.
【作者单位】：后勤工程学院信息工程系;
【基金】：国家自然科学基金(No.61271449,No.61302175) 重庆市自然科学基金(No.CSTC2011BA2015,No.CSTC2012jjA0877,No.CSTC2012jjA1047)
【分类号】：TN911.7

【参考文献】