利用小波去噪和HHT的模态参数识别

提出了基于小波去噪和HHT的模态参数识别方法,以改善模态参数识别的精度.该方法先利用小波进行信号去噪,克服噪声对EMD分解的影响,以减少EMD分解过程的计算量和分解层数,对去噪后的信号进行EMD分解提取单模态的自由响应,然后利用自由响应的Hilberr变换识别模态固有频率和阻尼比.利用该方法对某振动台简支梁系统进行了模态参数识别,结果表明在噪声干扰下,该方法识别模态参数的精度较高,特别是阻尼比识别精度高.     

适用于网壳结构的模态测试法及数值检验

提出了一种适用于网壳结构的模态测试方法.该方法联合使用单点瞬态冲击激励和复模态指示函数法,通过对激励位置的优化选取来分离密集模态,以减轻识别过程中密集模态间的混叠程度.复模态指示函数法是一种先进的频域模态识别方法,能够快速、自动地确定结构模态参数,优于传统的峰值检测法.文中对该方法进行了数值模拟,结果表明,该方法能够确保网壳结构模态参数识别的有效性和准确性.     

一种模态弱响应且模态密集的参数识别方法

针对运行模态分析中响应信号数据样本较短、模态密集,以及弱响应信号淹没在大噪声中,系统模态难以全部辨识的问题,提出了联合相关函数与传递率识别系统模态参数的方法(联合方法),该方法先应用传递率近似频响函数获取系统弱响应频率特征函数,然后再通过小波变换进行模态识别。运用随机激励下的GARTEUR飞机模型仿真运行状态的输出进行了数值仿真实验。结果表明:相比于多参考最小二乘复频域法,联合方法不仅能提高模态频率的识别精度,而且还能极大地提高阻尼比的识别精度,尤其是传递率对模态密集的弱响应模态识别结果良好。     

模糊聚类在正交多项式法模态参数识别中的应用

针对试验模态分析中密集模态识别精度差等问题,采用频率归一化正交多项式法进行多模态参数识别,并采用模糊聚类方法实现多测点极点的自动选择,减少识别过程中对分析人员经验的依赖。详细论述在频率归一化下系数矩阵正交化和转换矩阵的归纳求法以及模糊聚类方法的具体实现。仿真数据证明算法能够识别密集模态,剔除虚假模态,识别精度高。     

基于EEMD与盒维数的轴承故障特征提取

针对EMD存在的模态混叠以及轴承振动信号故障特征提取困难的问题,提出了一种基于EEMD与盒维数的轴承故障特征提取方法。首先,采用EEMD方法将去噪后的轴承振动信号自适应地分解为多个本征模函数。其次,利用相关系数和峭度选取一定量的本征模函数,并计算盒维数与能量熵得到特征向量。最后,将特征向量输入概率神经网络进行分类识别。仿真与实验结果表明在抗模态混叠以及轴承故障分类识别方面,该方法均优于EMD。     

基于平移不变小波阈值算法的经验模态分解方法

针对经验模态分解(EMD)处理存在间断事件的数据会引起模态混叠的问题,提出了一种经验模态分解与小波去噪相结合的新方法.由于传统的小波变换去噪会使信号在间断事件处产生人为的振荡现象,故先用平移不变小波去噪算法对信号进行预处理,消除间断信号对EMD方法的影响,再对信号进行EMD分解.理论计算及实验研究表明:只有在有效消除异常干扰的情况下才能获得可靠的IMF分量,采用本文的方法消除间断信号对EMD分解的影响是有效可行的.     

运营条件下异型拱桥的模态参数识别

以一复杂的异型钢管混凝土拱桥为对象，根据运营期间车辆交通荷载引起的余振实测结果，用小波变换（wavelet transform,简称WT）和特征系统实现法（eigensystem realization algorithm，简称ERA）两种方法对结构模态参数识别结果进行了对比，讨论了两种方法在复杂结构体系模态参数识别中的适用性。进一步以ERA识别方法为对象，比较了记录波除噪及不同记录波对参数识别结果的影响。结果表明：WT方法以及ETA方法得到的结构模态参数识别结果基本一致，两种方法相互校核可剔除噪声模态，获得真实的结构模态信息；除噪对基于余振衰减波的模态参数识别精度提高意义不大，不同记录波的对比识别可剔除单波随机噪声的影响，提高识别结果的可靠性。     

基于Hilbert-Huang变换与理想带通滤波器的系统识别

针对经验模态分解存在模态混叠现象,提出基于Hilbert-Huang变换与理想带通滤波器的系统识别方法。该方法利用傅里叶变换得到结构加速度响应频响函数,粗略估计固有频率范围,通过半功率带宽法设计理想带通滤波器,定量化确定通带带宽,使信号在经过滤波器后频域内零相移,同时不改变其幅值谱。结构响应通过指定频带的理想带通滤波器产生若干窄带信号,利用经验模态分解获取结构模态响应,经Hilbert变换构造模态响应解析信号,并通过线性最小二乘拟合提取结构模态参数与物理参数。结果表明:半功率带宽法可实现带通滤波器频带的定量化设计,理想带通滤波器的零相移特点较好契合Hilbert-Huang变换用于系统识别的要求,两者结合可有效地解决模态混叠现象,减少虚假模态,大大提高结构系统识别精度。     

基于EEMD-ICA去噪方法在转子故障特征识别中的应用研究

针对转子故障诊断过程中噪声干扰大,故障特征难以提取的问题,首次将基于集合经验模态分解-独立分量分析(EEMD-ICA)方法应用到转子故障特征识别中。该方法可有效抑制模态混叠现象,准确分离出包含在信号内部的噪声成分。实验结果表明,所提方法去噪效果明显,能够有效提取转子故障特征。     

基于小波去噪和EEMD包络解调分析的滚动轴承故障诊断方法

针对强噪声背景下的故障信号诊断问题,提出一种基于小波去噪和改进型总体经验模式分解算法(Ensemble Empirical Mode Decomposition,EEMD)包络解调分析的滚动轴承故障诊断方法。由于经验模态分解方法易产生虚假分量和模态混叠现象,引入EEMD。首先将采集到的振动信号进行软阈值去噪,然后对去噪信号进行EEMD分解,抽取能量较大的前4个内禀模态函数(IMF)进行Hilbert变换,得到包络信号,最后对包络信号进行细化谱分析,得到轴承故障特征频率。小波去噪可解决噪声造成的包络信号粗糙这一问题,提高了包络提取精度。将该方法应用于滚动轴承的内圈和外圈故障诊断,诊断结果均表明该方法能够准确有效地提取故障特征频率。     

Structural response reconstruction based on the modal superposition method in the presence of closely spaced modes

An approach for structural response reconstruction based on the modal superposition method in the presence of closely spaced modes is proposed in this paper. In this method, the entire mode-set of a structure is divided into closely spaced modes and the rest of the modes. The rest of the modal responses whose response is known will be separated into individual modal response by using the empirical mode decomposition method with intermittency criteria. Starting from the mode shapes, derived from these modal responses, the rest of the modal responses at the unavailable locations can be acquired. Furthermore, the contribution of the closely spaced modal responses at the unavailable locations can be obtained based on these responses. This proposed method is valid if there is no periodic excitation. However, in practice it is common that a structure might be excited by transient, stochastic, periodic forces or a combination thereof, i.e. hybrid excitations. A hybrid approach for solving the reconstruction problem of hybrid excitations is also developed, which is based on the proposed method and transmissibility concept in response reconstruction. Numerical studies are conducted and compared with theoretical predictions for validation. Effects of background noise level, high damping ratio and multiple forces are investigated in detail.     

Substraction technique and finite difference formulas for modal parameter estimation

This paper deals with the use of the substraction technique and the very simple finite difference formulas for modal parameter estimation. For closely spaced modes, the estimation of the formulas is not accurate and depends strongly on the frequency-response data involved in computations. An iterative procedure based on the substraction technique combined with the finite difference formulas is proposed. Application to experimental frequency response data shows that after the iterative procedure is applied, the estimation depends much less on the data involved, and the estimation reliability and accuracy are significantly improved, especially for closely spaced modes.     

基于奇异值分解的频响函数降噪方法

为了提高外场测试中频响函数的信噪比,提出了一种基于奇异值分解的频响函数降噪方法。该方法首先对脉冲响应函数在相空间内进行重构;然后对重构轨道矩阵进行奇异值分解达到降噪的目的。其中,降噪阶次通过奇异熵增量进行确定。采用GARTEUR飞机模型建立具有密集模态的仿真算例进行验证。结果表明,在噪声干扰较大时,该降噪方法能够显著改善模态参数的识别精度,尤其是阻尼的识别精度。     

PARAMETRIC TIME-DOMAIN METHODS FOR THE IDENTIFICATION OF VIBRATING STRUCTURES—A CRITICAL COMPARISON AND ASSESSMENT

A critical comparison of four stochastic (PEM, 2SLS, LMS, IV) and three deterministic (LS, Prony, ERA) methods for the parametric time-domain identification of vibrating structures from random excitation and noise-corrupted response signals is presented. Concise summaries of the methods, highlighting their principles and realisations, are provided, while the study is based upon a six-degree-of-freedom structural model characterised by two closely spaced modes, two weak modes and a wide range of modal damping. Monte-Carlo experiments under two different (wideband/narrowband) noise environments are performed, along with comparisons with non-parametric frequency domain identification.The stochastic methods—most notably PEM, LMS and IV—are, at the price of increased complexity, shown to lead to potential advantages in non-negligible noise cases, while deterministic methods—most notably Prony and ERA—may suffice under negligible noise. In addition: (a) Model order estimation is shown not to be straightforward, and significant overdetermination is required (especially by the LS). (b) A weak closely spaced mode is hard to identify, while being completely missed by the deterministic methods and the 2SLS. (c) A highly damped mode presents some difficulty as well (mainly for the LS). (d) False modes are exhibited, primarily by the LS (wideband noise) and the ERA (narrowband noise). (e) The achievable estimation accuracy is generally high for the natural frequencies, lower for the damping ratios, and even more so for the residues (mode shapes). Furthermore, accuracy is somewhat lower for the closely spaced modes and significantly lower for the two highly damped modes. (f) PEM, LMS and IV achieve lower bias errors and good overall accuracy, followed by the 2SLS, Prony, ERA and, finally, LS. (g) Unstable modes are mainly exhibited by the IV and ERA. (h) All methods appear sensitive to the selected model order and design parameters, and user expertise is necessary.     

A new theoretical basis for the bandwidth method and optimal power ratios for the damping estimation

Using the well-known bandwidth formula and the half power bandwidth formula [R.E.D. Bishop, G.M.L. Gladwell, An investigation into the theory of resonance testing, Philosophical Transactions of the Royal Society of London A 255 (1963) 241–280], in particular, is the simplest way to estimate modal damping for engineers. By using the half power bandwidth formula, the damping factor is estimated to be approximately the half bandwidth at the half power points. One of the major limitations that restrict the use of this method is the coupling effect between closely spaced modes. In this paper, the dependence of the damping estimation accuracy on the selected power ratios is studied with both analytical and experimental data of frequency response functions. The results show that by selecting adequate power ratio values, the coupling effect can be minimized and the estimation accuracy can be significantly improved for closely spaced modes. A further improvement of accuracy can be obtained by applying the algorithm of mode isolation [H.P. Yin, D. Duhamel, Substraction technique and finite difference formulas for modal parameter estimation, Mechanical Systems and Signal Processing 18 (2004) 1497–1503; M.S. Allen, J.H. Ginsberg, A global, single-input-multi-output (SIMO) implementation of the algorithm of mode isolation and application to analytical and experimental data, Mechanical Systems and Signal Processing 20 (2006) 1090–1111]. Also an exact bandwidth formula in case of a single degree of freedom system is presented and the link between the exact formula and the classical approximated formula is indicated. The exact bandwidth formula provides a new theoretical basis of the bandwidth method for the damping estimation from frequency response functions.     

Output-only analysis of structures with closely spaced poles

In the framework of the operational modal analysis, several approaches have been developed for estimating the modal parameters, i.e., natural frequencies, damping ratios, and mode shapes. Specifically, a technique capable to evaluate the biased (i.e., unscaled by a constant or an almost constant function) frequency response functions, FRFs, has been proposed. Assuming that only the responses of the structure are disposable, the technique allows one to estimate biased FRFs starting from the power spectral densities, PSDs, and applying the Hilbert transform. This paper deals with the estimates of the modal analysis parameters mentioned above. It is possible to obtain each single mode shape, from the singular vectors achieved by applying the singular value decomposition to the FRF matrix evaluated at the spectral line corresponding to the selected natural frequency. A special attention will be devoted to structures with coupled modes, i.e., closely spaced modes. Once the FRFs have been obtained, the natural frequencies and damping ratios could be achieved either in the frequency domain or in the time domain. Experimental tests, carried out on beams, plates and on the AB-204 helicopter blade, will be presented.     

结构密集模态非线性耦合作用引起的线性模型失效问题研究

通过对两自由度密频近线性系统主共振响应的分析,研究了结构密集模态弹性非线性耦合作用引起的一些典型的非线性动力学现象,如幅频曲线弯曲Ｈｏｐｆ分叉导致的概周期响应。研究结果表明,存在密频内共振时,线性模型失效,必须采用非线性模型来描述结构的动力学响应。     

柔性智能桁架结构的独立模态空间控制方法实验研究

针对空间柔性智能桁架结构 ,进行了实验室控制仿真。实验基于最优控制理论 ,采用独立模态空间控制方法设计了空间柔性智能桁架结构的振动主动控制规律。通过测量加速度 ,进行了柔性智能桁架结构的实时振动主动控制实验。实验结果表明该控制使被控模态阻尼比增加 ,频响函数的幅值衰减较大 ,有较明显的控制效果