EMD中异常事件处理及其在转子动平衡中的应用

Abnormal events processing of empirical mode decomposition and its application in rotor dynamic balancing

经验模态分解(EMD)广泛用于将非平稳、非线性信号分解成一组稳态和线性的数据序列集,得到理论意义上仅含有单组分频率成分的各固有模态函数。为减少异常事件对信号EMD分解的干扰,提出采用相关分析消除信号中异常事件的方法。同时结合不平衡振动理论提出一种准确分离转子不平衡振动信号的方法,以用于转子不平衡振动监测及动平衡等方面。以实例分析证实该方法的有效性。

The Empirical Mode Decomposition （EMD） is used for decomposing nonstationary and nonlinear signals to stationary and linear ones, and Intrinsic Mode Functions （IMFs） can be gained, they contain only a single group of frequencies. In order to reduce interference of abnormal events to the EMD method, the correlation analysis was used to eliminate abnormal events. The method to separate rotor imbalance vibration signals accurately was put forward based on EMD method and imbalance vibration theories, it could be used to monitor imbalance vibration and dynamic balancing. The validity of the proposed method was verified with example analysis.