模态参数识别的LMD-BPF方法及其应用

LMD-BPF method for modal parameter identification and its applications

鉴于准确识别结构模态参数存在的问题,提出了一种局部均值分解(LMD)-带通滤波(BPF)的模态参数识别方法。首先,将LMD方法用于位移仿真信号和压气机导向叶片测频信号的分解,分解得到的PF分量存在模态混叠现象;然后,使用LMD-BPF方法对位移仿真信号和导向叶片测频信号进行分解,成功实现了对各模态频率的准确分离;最后,使用LMD-BPF模态参数识别方法对位移仿真信号和导向叶片测频信号进行模态参数识别。识别的位移仿真信号4个模态频率和阻尼比与相应的理论值之间的最大误差分别为0.205%和2.387%,识别导向叶片的3种模态频率与测试模态频率之间的最大差别小于1.0%,识别的导向叶片的3种模态阻尼比与半功率带宽法识别的阻尼比,最大差别小于0.65%。模态参数识别的仿真分析和实验研究验证了该方法的有效性。

In view of the difficulties existing in the accurate identification of structures’modal parameters,a modal parameter identification method based on local mean decomposition(LMD)and band-pass filtering(BPF)was proposed.Firstly,the LMD method was used in the decomposition of a displacement simulation signal and a frequency measuring signal of a compressor guide vane,and modal aliasing phenomenon existed in the decomposed PF components;Then,the LMD-BPF method was utilized in the decomposition of the displacement simulation signal and frequency testing signal of the guide vane,and each modal frequency was accurately and successfully separated;Finally,the LMD-BPF modal parameter identification method was employed to identify the modal parameters of the displacement simulation signal and frequency testing signal of the guide vane.The maximum errors between the four modal frequencies and damping ratios of the identified displacement simulation signal and the corresponding theoretical values were 0.205%and 2.387%,respectively.The maximum difference between the three modal frequencies of the identified guide vane and the testing frequencies was less than 1.0%,and the maximum difference between the three modal damping ratios of the identified guide vane and those identified by half power bandwidth method was less than 0.65%.The simulation analysis and experimental study of modal parameter identification verified the validity of the proposed method.