【摘要】 基于ARMA时间序列分析模型,建立了一种在部分振动信号缺失情况下进行预测恢复的方法。借助于MATLAB系统辨识工具箱,对振动信号时间序列进行了相关分析、模型参数的估计及检验。最后,将经过检验的ARMA模型应用于某稳压泵实测振动信号中缺失数据的预测恢复。更多还原

【Abstract】 Based on the auto-regressive and moving average (ARMA) model,a method for recovering vibration missing data is developed. By using MATLAB system toolbox,the correlation analysis,model parameter estimation and inspection of vibration time-series signal are executed. The ARMA-model method is applied to the pump vibration data to recover the missing data. The results show that the method is feasible and effective in engineering application.更多还原