基于小波域AR模型的直升机结构模态参数辨识  

Modal parameters identification of a helicopter based on AR model in wavelet domain

作　　者：何绪飞[1] 邓忠民[1] 武晓东[1] 

机构地区：[1]北京航空航天大学宇航学院

出　　处：《振动与冲击》2010年第9期201-205,共5页Journal of Vibration and Shock

基　　金：国家自然科学基金(10972019)

摘　　要：小波变换具有良好的时频分辨率,在结构动特性参数辨识领域引起了广泛的关注。利用连续小波变换,给出一种小波域的模态参数识别算法。对某型号直升机机身的时序响应进行分析,使用Coiflets小波基函数建立小波域的自回归(AR)模型。模型系数决定了系统的动力学特性,利用伪逆法求解模型的特征矩阵,系统模态参数的辨识问题就转化为求解特征矩阵的特征值和特征向量。讨论了小波变换尺度、时移参数及噪声对模态参数辨识精度的影响。将模态识别结果分别与有限元分析及其它方法所得结果进行比较,表明所用小波变换方法分辨率高、抗干扰能力强,适用于直升机结构的模态参数辨识。A wavelet-based approach was provided to identify the modal parameters of a certain type of helicopter.An auto regression(AR) model in wavelet domain was established by applying a continuous wavelet transform to the time-history responses of the fuselage.The Coiflets function was taken as the mother wavelet.The coefficient matrices of the AR model,which parmeterize the dynamic characteristics of the system,were calculated by using the pseudo-incerse method.Modal parameters were further obtained by solving the eigenvalue matrix.The effects of wavelet transform scaling coefficient,shifting parameter and noise on the ability to accurately evaluate the modal parameters were investigated.The results comparison was carried out with finite element analysis and other methods,which shows that the applied wavelet method is of high resolution and insensiticve to noise,and suitable for modal parameters identification of helicopters.

关 键 词：小波变换 模态分析 动力学响应 参数识别 自回归 

分 类 号：TU311.1[建筑科学—结构工程] V214.1[航空宇航科学与技术—航空宇航推进理论与工程]