一种新的信号特征量检测方法及其在自动平衡中的应用研究

快速准确获得振动信号的相位、幅值二特征量具有重要工程意义。本文提出了一种无需整周期采样、无需锁相环和复杂电路硬件的相位检测算法。该算法基于互功率谱估计,具有算法复杂度低、精确度高、适用频率范围广等优点。在该算法基础上又提出了一种相位补偿滤波算法,使信号经滤波后相位保持不变。该方法的运算量低,适于在线实时滤波;相位补偿准确度高,误差在1°以内。综合上述两算法构成的信号特征量检测方法,可以准确、快速地提取振动信号的相位及幅值。最后本文将上述方法应用于旋转机械自动平衡,结果表明两算法可以在线、准确地实现,并且能够达到较好的平衡效果。

Study of a Novel Approach to Signal Characteristic Quantity Detection and its Application in Automatic Balance

Quickly acquiring characteristic quantities such as the exact phase and amplitude of vibration signal is significant in engineering. A novel phase detection algorithm that does not need complete period sampling, phase-locked loops or complex electrocircuit hardware was presented in this paper.Based on the cross-power spectrum estimation,the algorithm has the advantages of lower complexity and better precision over wide frequency band.With this algorithm,a new phase compensation filter algorithm is developed for realizing zero phase shift.With less operation burden and good phase compensation accuracy(error less than one degree),it is suitable for real-time online filtering.When the aforementioned two algorithms are combined,a novel approach to detecting signal characteristic quantity,whereby the phase and amplitude of vibration signals can be extracted exactly and quickly,will come into being.The approach is used in the automatic balance of rotational machines.The result shows that it can realize online accuracy and perfect automatic balance.