振动信号频率跟踪的反馈修正自适应陷波器法

Feedback corrected adaptive notch filter for vibration signal frequency tracking

采用自适应陷波器跟踪振动信号频率时,存在持续跟踪精度不高的问题。通过分析指出持续跟踪精度不高的根本原因是ANF误差收敛至局部最优值,借鉴反馈控制思想,提出一种反馈修正ANF。根据ANF输入输出信号的相关性,设计频率跟踪精度评估因子,实时监控ANF频率跟踪精度。若评估因子小于给定阈值,则认为ANF丢失振动信号频率,通过反馈调整ANF参数而非重新初始化来适当增加陷波带宽,使其能重新跟踪到信号频率又具有较快的重新收敛速度。以一种基于Steiglitz-McBride方法的ANF(SMM-ANF)为例,分析了反馈修正策略,给出流程和具体算法。仿真比较了格型ANF、SMM-ANF和反馈修正SMM-ANF的性能,给出科里奥利质量流量计应用实例,结果表明:反馈修正SMM-ANF收敛速度稍慢于SMM-ANF,快于格型ANF,持续跟踪精度明显提高。

The precision of vibration frequency estimation with an adaptive notch filter （ANF）is not satisfactory all the time.It was found through analysis that the fundamental reason is the ANF error converges to a local optimum value. Here,a feedback corrected ANF was proposed of an ANF,with the idea of feedback control.According to the correlation between an input signal and an output one,a factor evaluating its frequency tracking accuracy was designed to monitor its working status.If the factor was less than a given threshold value,it was said that the ANF lost vibration frequency. Then,it was demanded to enlarge the bandwidth of the ANF so that the frequency could be re-tracked accurately.To insure the faster re-converging speed,a feedback strategy adjusting the ANF＇s parameters appropriately rather than re-initialization was applied.An ANF based on Steiglitz-McBride method（SMM-ANF）was taken as an example to illustrate the feedback correction strategy.Its flow process and the corresponding algorithm were also exhibited.The performances of a Lattice ANF,a SMM-ANF and a feedback corrected SMM-ANF were compared with simulations.An instance of the proposed method applied in Coriolis mass flow meter was given briefly.Results showed that the convergence rate of the feedback corrected SMM-ANF is lower than that of the original SMM-ANF,higher than that of the lattice ANF;its continual frequency tracking performance is obviously better than that of the others.