应用优化的同伦分析法求解非线性Jerk方程

应用优化的同伦分析法计算了具有三次非线性项的三阶微分方程(Jerk)的近似周期和近似解析周期解。文中给出一个算例说明由优化的同伦分析法可以容易得到精确的二阶近似周期解。当初速度 比较大时,一阶近似周期与精确周期的百分比误差为-0.415%,而二阶近似周期与精确周期的百分比误差为-0.0298%。与数值方法给出的“精确”周期解比较,一阶近似解析周期解和二阶近似周期解的精度很高。这个说明同伦分析法对求解非线性Jerk方程非常有效     

退化环面上的非线性jerk方程近似周期解的同伦分析方法

用同伦分析法求解退化环面上的非线性Jerk方程的近似周期和近似解析周期解。所得结果表明文中得到的一阶近似周期和一阶近似解析周期解与Gottlieb用低阶谐波平衡法求解得到的结果一样。当参数和初速度较大时,一阶近似周期与精确周期的百分比误差是4.831 8%,而二阶近似周期与精确周期的百分比误差小于0.219 9%。与数值方法给出的"精确"周期解比较,二阶近似解析周期解比一阶近似解析周期解要精确的多。因此,同伦分析法是求解非线性Jerk方程的一种非常有效的方法。     

基于损伤存在概率成像方法的复合材料结构损伤识别

基于小波分析理论和概率统计原理,提出一种损伤存在概率成像方法,对复合材料结构进行在线的健康监测。首先对结构损伤前后Lamb信号进行比较,提取信号的能量特征差异系数,作为损伤指标;然后用概率统计方法判断该损伤指标是由损伤引起的还是因环境变化造成的;最后用成像算法给出存在概率图像识别损伤。对复合材料板结构进行实验验证该方法的可行性,具有一定的实际工程应用价值。     

求解非线性互补问题的单纯同伦算法

在更弱的情况下,我们首先给出了非线性互补问题存在解的几个充分条件,然后给出一个求解的单纯同伦方法。     

一类非线性问题的同伦解法

利用同伦映射理论,本文研究了一类非线性问题。在适当的条件下较简捷地得到了问题解的估计式,并对一些典型的问题,解的估计式达到了较高的精度。     

基于提升小波熵指标的梁板组合桥损伤识别研究

利用提升小波和熵的特点,提出了基于提升小波熵的结构损伤识别方法,给出了提升小波时间熵和相对提升小波熵两个结构损伤指标。通过梁板组合桥的数值模拟表明,提升小波时间熵指标能较好地识别结构的损伤时刻,而相对提升小波熵指标能较好地识别结构的损伤位置。     

干摩擦自激振动周期解的同伦方法

针对一类单质体-传输带干摩擦系统,首先对系统平衡点进行稳定性分析,然后利用同伦方法研究系统的纯滑动和粘滞滑动形式的干摩擦自激振动。通过适当选取基函数,定义初始猜测解、辅助线性算子和非线性算子,计算到二阶或三阶近似即可得到精度较高的解析近似解。由于粘-滑极限环的由两个阶段组成,具有非光滑特点,在计算其滑动阶段时,可以允许近似解中出现长期项。     

基于时间序列分析的结构损伤识别

针对如何从结构响应信息中提取结构损伤指标的问题,提出了一种基于时间序列分析的结构损伤识别方法。对结构响应数据进行预处理,利用完好工况下的结构响应数据作为参考数据,建立自回归(Auto regression-AR)预测参考模型。利用已建立的AR预测参考模型计算待识别工况的残差,将待识别工况的残差与AR预测参考模型的残差的方差之比作为损伤指标,对结构损伤进行识别;通过算例表明:该损伤指标不仅可以判断结构是否发生损伤,而且可以识别结构的损伤位置。     

基于单元残余能量灵敏度的结构损伤识别

在定义单元残余能量基础上进行单元残余能量的灵敏度分析,提出基于单元残余能量灵敏度分析的结构损伤识别方法,建立结构损伤识别方程组。通过解方程,求得损伤系数向量,便可同时识别结构的损伤位置及损伤程度。仿真结果表明,本文提出的方法对各损伤工况不但能准确定位,而且能有效识别出损伤程度,识别精度较高。     

基于模型修正的钢管焊接结构焊缝损伤识别

摘要：本文提出了一种基于模型修正的钢管焊接结构焊缝损伤识别方法。利用从发射台骨架试验模型获取的模态参数,选择识别结果中精度较好的模态频率作为模型修正的基准频率。通过对待修正参数的灵敏度分析,运用ANSYS和MATLAB软件对有限元模型进行了修正。以实测模态和计算模态之间的误差建立一个带约束边界非线性最小二乘目标函数,将损伤识别问题转化为优化问题,并采用信赖域方法求解该优化问题。以有限元模型焊接结点单元组弹性模量的降低模拟焊缝损伤,并假定了两种损伤工况,通过对发射台骨架模型的数值仿真及试验研究,结果表明本文提出的损伤识别方法识别效果较为理想,为解决这种复杂焊接结构焊缝损伤识别问题提供了新的思路。     

基于EEMD与FastICA的损伤异常识别与定位

为了准确地提取结构损伤异常信息,消除小波奇异值分解时存在需要特定的小波基和分解层数以及经验模态分解（EMD）方法存在诸如虚假模态混叠等问题,本文提出了一种基于改进的总体平均经验模态分解（EEMD）与快速独立分量分析（FastICA）相结合的提取结构损伤特征并进行识别与定位的新方法。首先,通过EEMD对结构动力响应信号进行预处理并用 FastICA提取出包含损伤信息的特征分量对结构响应异常进行识别和初步定位;然后,计算归一化的源分布向量(NSDV)的最大值,并根据该最大值精确定位结构损伤。最后,通过框架数值算例和试验进行了所提方法的验证,结果表明该算法能够较好地进行结构损伤异常的识别与定位。     

Application of mixed variational formulations based on the finite element method to the solution of problems on natural vibrations of elastic bodies

We consider mixed variational formulations and the application of the mixed approximations of the finite element method to the solution of problems on natural vibrations of elastic bodies. To solve the generalized spectral problem, three forms of the mixed variational formulations are proposed. The correctness and stability of mixed variational formulations for displacements, strains and stresses are investigated. Matrix equations of the mixed method are given whose solution is performed using the modified algorithm of the steepest descent method. The results of calculations for natural frequencies of free vibrations of a straight and a circular beam are presented that are obtained in the solution of the problem in a two-dimensional formulation based on the classical and mixed finite-element method approaches. __________ Translated from Problemy Prochnosti, No. 2, pp. 121–140, March–April, 2008.     

SAW filters based on parallel-connected coupled resonator filters with offset frequencies

The concept of coupled resonators is applied to synthesize surface acoustic wave filters. Employing two parallel-connected filter tracks, with a frequency shift imposed between them, a wide passband with low insertion loss together with well-controlled rejections is achieved. The operation of the two-track device is based on the mutual interaction of the individual transfer functions for the pair of tracks. Each track serves to contribute a part of the passband, enabling a wide band. Outside of the passband, the signals passing through the two channels may cancel each other, thus facilitating efficient control over the rejections. However, obtaining rejection stopbands at just the predetermined frequencies requires precise values for the materials parameters and a reliable fabrication process. Prototype devices fabricated with this approach are demonstrated both on quartz and, for the first time, on 42 degrees-LiTaO3. Results for two-track devices having either two or three transducers per track and operating either single-ended or with a balanced output are presented. The devices are designed employing the coupling-of-modes model and transmission-matrix approach, and the separate tracks are optimized simultaneously and independently. The center frequencies are 868 MHz and 1960 MHz. On quartz, a minimum insertion loss of 4 dB and a passband width of 0.23% are achieved at 868 MHz. On 42 degrees-LiTaO3, the corresponding figures of merit are 1.3 dB for minimum insertion loss and 4.1% bandwidth at 1960 MHz. The filters on 42 degrees-LiTaO3 also have remarkably flat passbands.     

基于分布式应变模态残差统计趋势的大跨斜拉桥损伤识别策略

近几年新兴的分布式测试技术得到了飞速发展,使得基于这类应变测试数据的损伤识别方法的研发显得尤为迫切。本文首先引入了分布式应变模态差损伤指标,并利用此指标对一大跨斜拉桥进行了损伤识别研究,分析了不同的测试环境、随机激励及损伤程度等因素对损伤识别效果的影响。然后对同种健康状态下的结构在不同测试环境条件下进行多次辨识,利用统计趋势分析和概率置信准则,提出了基于准分布式应变模态残差统计趋势的损伤识别策略。数值计算表明：所提策略很好的降低了各种随机干扰的不利影响,对斜拉桥主箱梁上发生损伤的多处单元都能准确识别出来,很好的解决了传统模态法损伤指标对结构早期轻微损伤不敏感的缺陷,避免了损伤淹没和误判现象。     

Crack detection by modal analysis in 3D beams based on FEM

During the past decade, the development of Structural Health Monitoring (SHM) systems has gained considerable interest in several engineering fields. The possibility of monitoring structure status under various operational conditions has resulted in a drastic change in the maintenance schemes. The introduction of SHM systems would allow a transition of the maintenance strategy from scheduled basis (i.e. Time Based) to a condition basis. The development of SHM systems relies on the availability of reliable techniques to acquire the characteristic features of damage through experimental measurements. In this paper, the flexural vibration of a cantilever beam, fixed-roller beam, and portal frame with a transverse surface crack is considered. The natural frequencies and mode shapes are computed by the Finite Element Method (FEM) using SAP2000 Software for various crack locations and depths. The computed mode shapes are animated to give the designers a good picture on the failure modes and shapes. The modal frequencies are experimentally measured and compared with the computed values and showed good compatibility. The sensitivity of the modal frequencies to a crack increases when the crack is near the fixed end, and decreases as the crack moves towards the free end of the beam or frame. The present investigations have revealed the following generalized results. The presence of crack in structure leads to an appreciable difference in the dynamic response. The findings based on condition monitoring technique can be utilized in various industrial applications, particularly for fault detection on structures. The proposed technique represents actually a non-destructive procedure having great benefits for health monitoring of structures.     

Electrogenerated chemiluminescence detection of amino acids based on precolumn derivatization coupled with capillary electrophoresis separation

A novel method for highly sensitive detection of primary and secondary amino acids with selective derivatization using acetaldehyde as a new derivatization reagent was proposed by capillary electrophoresis (CE) coupled with electrogenerated chemiluminescence (ECL) of tris(2,2'-bipyridine)ruthenium(II). The precolumn derivatization of these amino acids with acetaldehyde was performed in aqueous solution at room temperature for 1 h. Upon optimized derivatization, the ECL intensities and detection sensitivities of the amino acids were significantly enhanced by 20-70 times. Using four amino acids, arginine, proline, valine, and leucine, as model compounds, their derivatives could be completely separated by CE and sensitively detected by ECL within 22 min. The linear ranges were 0.5-100 microM for arginine and proline and 5-1000 microM for valine and leucine with the detection limits of 1 x 10(-7) (0.5 fmol, arginine), 8 x 10(-8) (0.4 fmol, proline), 1 x 10(-6) (5 fmol, valine), and 1.6 x 10(-6) M (8 fmol, leucine) at a signal-to-noise ratio of 3. The derivatization reactions and ECL process of amino acids were also proposed based on in situ Fourier transform infrared and ultraviolet spectrometric analyses.     

Mechanical fatigue in aluminium at elevated temperature and remaining life prediction based on natural frequency evolution

Fatigue is one of the principle damage mechanisms in materials especially at elevated temperatures. It possesses a great influence on the material properties and the possible working life. In this work, a novel technique for the analysis of fatigue at elevated temperature is suggested. Material fundamental frequency and its drop are used as the key parameters to predict the remaining useful life of a selected shape specimen, which is operated under low‐cyclic fatigue loading at elevated temperature. Experiments are performed on aluminium 1050 with two different lengths of a non‐prismatic cantilever beam at room temperature, 100, 200 and 300 °C, respectively. The experimental data is further transformed in to an empirical correlation that can predict specimen useful life. The error in this life prediction is less than 5%, and it reduces more once the frequency drop is increased.     

Phase I outlier detection in profiles with binary data based on penalized likelihood

Profile monitoring has been proven to be important among statistical process control problems. In some specific applications, the response variable of a profile can be categorical data, and numerous methods have been proposed for monitoring this type of profiles. Nevertheless, outlier detection for profiles with categorical data still attracted insufficient attention in the literature. To this end, this paper focuses on binary profiles and develops two schemes for outlier detection, from the viewpoint of penalized likelihood, based on the group LASSO method and directional information, respectively, in which the profiles of interest are treated as an integrated high‐dimensional vector. Simulation study shows that the proposed group‐type scheme usually performs better than the existing chart in terms of detecting outliers correctly and alleviating the masking effect. Finally, a real example is used for illustrating the implementation of the proposed scheme.     

Effect of model ligands on iron redox speciation in natural waters using flow injection with luminol chemiluminescence detection

The effects of dissolved organic compounds on the determination of nanomolar concentrations of Fe(II) have been compared using two luminol-based flow injection chemiluminescence (FI-CL) methods. One used the direct injection of sample into the luminol reagent stream, and the other incorporated on-line solid-phase extraction of the analyte on an 8-hydroxyquinoline microcolumn. The CL signals from analyses of dissolved iron species (Fe(II) and Fe(III)) with model ligands and organic compounds were examined in high-purity water and seawater. The organic compounds included natural reducing agents (e.g., ascorbic acid), nitrogen sigma-donor/pi-acceptor compounds (e.g., 1,4-dipyridine, protoporphyrin IX), aromatic compounds (e.g., 1,4-dihydroxybenzene), synthetic iron chelators (e.g., EDTA), and natural iron binding compounds (e.g., desferrioxamine B, ferrichrome A). Fe(II) determinations for both luminol FI-CL methods were affected by submicromolar concentrations of redox-active compounds, strong iron binding ligands (i.e., log K(FeL) > 6), and compounds with electron-donating functional groups in both high-purity water and seawater. This was due to reactions between organic molecules and iron species before and during analysis, rather than chemiluminescence caused by the individual organic compounds. In addition, the effects of strong ligands and size speciation on Fe(II) recoveries from seawater following acidification (pH 2) and reduction (100 microM sodium sulfite) were investigated.     

基于飞秒激光与纳秒激光协同多光子诱导荧光的火焰中氢原子瞬态检测

为了准确测量碳氢燃料燃烧过程中原位氢原子,提出了飞秒-纳秒激光协同激发的氢原子瞬态在线探测策略。以243 nm的飞秒激光经过双光子过程将基态氢原子激发到2S能级,随后这些氢原子再被656 nm的纳秒激光激发到3P能级,通过探测3P-2S的荧光发射,实现了宽当量比范围的氢原子无干扰瞬态在线测量。实验结果表明：通过飞秒-纳秒协同的方式,可以有效降低激光光解产生的氢原子对原位氢原子探测的干扰。