Kalman滤波在机械故障信号预处理中的应用

The Application of Kalman Filter on the Processing of Mechanical Fault Signals

简要介绍了Kalm an滤波的基本原理及其递推算法,给出利用时间序列自回归(AR)模型建立信号状态方程和观测方程的方法,并依托这两个方程结合Kalm an递推算法形成一种信号滤波处理方法,最后通过仿真和实测信号对这一方法进行验证。研究结果表明:利用Levinson-Durb in算法获得AR系数建立信号模型,再通过Kalm an递推能够有效地剔除振动信号噪声。

The essential principle and recursion algorithm of Kalman filter were introduced and the methods of establishing state and measurement equation based on time series auto recursive model were given out. A new signal filtering way was proposed according to Kalnlan recursion algorithm and the equations. Meanwhile this method was tested by simulation and practical signal. The research shows that it is very effective to remove the noise from the vibration signal by using Kalman filter on the condition of the signal model that is created bv using AR coefficient coming from Levinson -Durbin algorithm.