基于阶次谱的Vold-Kalman滤波带宽优选方法

发布时间：2018-06-04 19:54

本文选题：故障诊断 + Vold-Kalman滤波　；参考：《振动工程学报》2017年02期

【摘要】：Vold-Kalman滤波阶次跟踪分析技术以其能够在时域中提取阶次分量,同时避免了传统计算阶次分析的重采样和时、阶次域变换带来的误差等优点,已经成为了一种重要的非平稳状态下旋转机械的故障诊断方法。滤波带宽作为Vold-Kalman滤波阶次跟踪分析技术实现过程的核心参数,决定了该方法分析结果的准确性和稳定性,然而目前针对滤波带宽的选择主要依靠实践经验确定,严重制约了该方法的深入研究和实践应用。因此提出了一种基于阶次谱的Vold-Kalman滤波带宽优选方法,该方法结合了阶次数据的时域、阶次域的特点,并利用标准差指标,实现监测过程可控的Vold-Kalman滤波阶次跟踪分析。该方法可以根据不同故障类型以及使用者精度要求对滤波带宽进行适时更新,具有很大的应用价值。最后通过实验数据说明了该方法的实现过程及其有效性。
[Abstract]:Vold-Kalman filter order tracking analysis technology can extract order component in time domain, and avoid the advantages of traditional order analysis resampling and time domain transformation. It has become an important fault diagnosis method for rotating machinery under non stationary state. Filter bandwidth is used as Vold-Kalman filter. The core parameters of the order tracking analysis technology determine the accuracy and stability of the analysis results. However, at present, the selection of filter bandwidth depends mainly on practical experience, which seriously restricts the in-depth study and practical application of the method. Therefore, a Vold-Kalman filter bandwidth advantage based on order spectrum is proposed. The method combines the characteristics of the time domain and order domain of the order data, and uses the standard deviation index to realize the Vold-Kalman filtering order tracking analysis of the monitoring process. This method can update the filter bandwidth according to the different fault types and the user's precision requirements. The method is of great application value. Finally, the method is of great value. Experimental data illustrate the implementation process and effectiveness of the method.
【作者单位】：电子科技大学机械电子工程学院设备监控与健康管理研究所;
【基金】：国家自然科学基金资助项目(51305067,51375078)
【分类号】：TH17

【相似文献】