基于变分模态分解和同步提取变换识别时变结构瞬时频率

Instantaneous frequency identification of time-varying structures using variational mode decomposition and synchroextracting transform

针对同步提取变换(SET)不能分离频率成分间隔相近的多分量信号的问题,提出了一种结合变分模态分解(VMD)和同步提取变换识别时变结构瞬时频率的方法。首先,通过傅里叶变换确定预设模态数量,利用VMD对多分量信号进行分解得到多个模态分量;然后,采用SET对每个模态分量进行时频分析获取瞬时频率;最后,将各模态分量的时频谱图叠加得到完整的多分量信号时频谱图。针对多分量时变信号和两自由度时变结构自由振动响应信号的瞬时频率识别结果,验证了基于VMD和SET结合方法识别时变结构瞬时频率的有效性和正确性。结果表明,该方法具有较好的噪声鲁棒性和能量聚集性,克服了SET处理频率成分间隔相近的多分量信号的不足,能有效识别具有近距离频率成分的时变结构瞬时频率。索力线性和正弦变化时拉索瞬时频率识别的试验验证了该方法的适用性。

Aiming at the problem that synchroextracting transform(SET)cannot separate multi-component signals whose frequency components are close to each other,a method combining variational mode decomposition(VMD)and synchroextracting transform was proposed to identify the instantaneous frequency of time-varying structures.Firstly,the number of preset modes was determined by the Fourier transform,and the multi-component signal was decomposed by using VMD to obtain multiple mode components.Then,SET was used to analyze each modal component to obtain the instantaneous frequency.Finally,the time-frequency representation of each modal component was superimposed to obtain a complete time-frequency representation of the multi-component signal.The results of instantaneous frequency identification for multi-component time-varying signals and two degree of freedom time-varying structural free vibration response signals verified the effectiveness and correctness of the method based on VMD and SET.The results show that the method has strong noise robustness and energy aggregation,It circumvents the disadvantage of the SET that cannot process multi-component signals with close frequency components to effectively identify the instantaneous frequency of time-varying structures with close frequency components.The applicability of the method was verified by the experiment of identifying the instantaneous frequency of a cable when the cable force changes linearly and sinusoidally.