Buckling of composite beams with two non-overlapping delaminations: Lower and upper bounds

In this paper, lower and upper bounds of the buckling load of a composite beam with two non-overlapping delaminations are obtained by developing analytical models. The characteristic equation governing the delamination buckling is derived by using Euler-Bernoulli beam theory, performing proper linearization and by imposing the appropriate continuity and boundary conditions. The effects of the differential stretching and the bending-extension coupling are considered. The accuracy of the approach is verified by comparing results with previously published data and a separately carried out finite element analysis. The effects of the dimensions of the delaminations, their thicknesswise and spanwise locations on the lower and upper bounds of the buckling load are investigated in detail. The longer of the two non-overlapping delaminations dominates the buckling behavior. Composite beams with anti-symmetric non-overlapping delaminations withstand lower buckling loads than the composite beams with symmetric non-overlapping delaminations. The lower and upper bounds of the buckling load will be useful to gauge the working range of bridging and give guidelines for practical applications.     

Computational simulations of delamination wear in a coating system

Delamination wear constitutes one relevant wear process in many materials. In particular, wear failure through delamination is relevant for material systems where a coating is present on a substrate material. A computational modeling approach is presented that aims to describe the processes of formation and growth of wear delaminations. The model is based on the cohesive zone model approach. An irreversible cohesive zone model is used in a parametric numerical study of delamination wear in a coating system. The model predicts trends in agreement with the trends emerging from Archard's law. The proposed modeling approach has the advantage that details of the delamination wear process can numerically be studied, and that a unified framework from delamination initiation and propagation is provided.     

Effect of temperature on stresses and delamination failure of z-pinned joints

Although z-pins have been shown effective in preventing delaminations in adhesively bonded and co-cured joints, their applicability depends on a reliable assessment of the strength of a z-pin-composite assembly. In particular, high residual thermal stresses that have been found in experiments dictate the necessity in a local stress analysis. Elevated temperature applied to the joint during its lifetime may also affect its effectiveness in preventing delaminations. The present paper illustrates an approach to determining local residual stresses confirming the observations regarding a possible delamination and cracking in the composite structure due to high post-processing transverse stresses. The analysis of the effect of elevated temperature applied at one of the surfaces on the response of a z-pinned joint is conducted using the concept of a double cantilever beam with an “insulated” crack. In addition, it is illustrated that an elevated temperature may actually benefit the integrity of the joint if it causes an increase in the z-pin-composite interfacial strength.     

基于小波变换和ICA特征提取的开关电流电路故障诊断

提出了采用小波变换和独立成分分析(ICA)作为预处理器来进行特征提取的神经网络开关电流电路故障诊断方法。该方法对采集到的故障响应信号进行Haar小波正交滤波器分解,获得低频近似信息和高频细节信息;然后利用独立成分分析方法进行ICA故障特征提取;最后将所得到的最优故障特征输入到BP神经网络中进行故障分类。对六阶切比雪夫低通滤波器和六阶椭圆带通滤波器电路进行了仿真实验验证,获得了100%的故障诊断准确率,与其他方法进行比较,实验结果显示了该方法的优越性。     

Buckling analysis of a laminated composite plate with delaminations using the enhanced assumed strain solid shell element

The paper deals with the validation of a recently proposed hexahedral solid-shell finite element in the buckling analysis of a laminated composite plate with delaminations. The object is to study the buckling behavior of structures with delaminations using the enhanced assumed strain (EAS) solid shell element with 5, 7 and 9 parameters. The EAS three-dimensional finite element formulation presented in this paper is free from shear locking and leads to accurate results for distorted element shapes. The developed FE model is used to study the effects of some parameters in the buckling load, such as the stacking sequences, delamination size, aspect ratio, width-to-thickness ratio. The feasibility of the proposed method is confirmed by numerical examples. Results show that using hexahedral solid-shell finite element in the buckling analysis is more efficient than using the enhanced solid finite element.     

滚动轴承缺陷振动建模及冲击特征提取

以SKF6205-2RS深沟球轴承为研究对象,运用Hertzian接触理论、弹性力学及滚动轴承几何学,建立了可预测滚动轴承不同损伤位置和程度的状态模型,并通过Runge-Kutta数值方法获取了系统响应。计算结果表明:径向负荷作用下,模型对内圈、外圈、钢球局部损伤所激励的频率及其谐波成分可进行良好预测;阐述不同状态下轴承振动规律,论证连续Haar小波变换周期性;利用连续Haar小波变换在时间-尺度域上所特有的周期性结合自相关消噪,提出了一种滚动轴承早期损伤特征提取的自相关连续Haar小波方法。诊断实例证明,这种方法能够有效消除信号的噪声,提取信号的弱冲击成分。     

Eddy-Current Nondestructive Detection of Delaminations Near Edges of Composite Materials

The paper reports on an investigation of the field generated by an eddy-current transducer placed over an edge surface of composite material with delaminations. The calculations have been performed by the finite difference method, and modulation characteristics of vector potential components have been determined. The effect of flaws has been determined on the complex plane of the vector potential at different values of the generalized parameter , and the effect of the delamination width has been studied.     

Study on low frequency energy harvesting system in laminated aluminum beam structures with delamination

Piezoelectric material based energy harvesting system (EHS) has been designed and developed to study the influence of delamination on EHS in low frequency vibrating beam structures. The stiffness reduction due to the presence of delamination of specified size and its location is evaluated using the harvested energy in different vibrating elastic modes. Four aluminium beam specimens were fabricated, each having two layers of 1 mm thickness and a specified dimension of delamination was introduced by using a thin Teflon film. The beam without delamination is considered to be healthy and served as the reference structure. The PZT-5H patches were surface bonded on the beams to work as energy harvesting transducers. A suitable electronic circuit is developed to receive the energy from the vibrating beam. The experimental results in terms of energy generated by the four laminated beams have been compared with analytical results at resonant conditions and correlated to assess the effect of delamination on energy produced by low frequency vibration modes. The location of delamination has made a significant change in the harvested energy of second and third bendings; in particular, the delaminations at edge and root have displayed a clear trend. Resonant and off-resonant excitations have revealed that the edge delaminated beam produces lower energy output. The current work has demonstrated that energy harvesting from different elastic modes and with a variable frequency at constant force excitation can be a useful health monitoring technique, employing low frequency vibration, besides utilising the harvested energy itself.     

Post-buckling and delamination propagation behavior of composite laminates with embedded delamination

Delamination occurred due to poor manufacturing process or in-service actions significantly affects the mechanical and failure behavior of laminated composite structures. In this study, the buckling and post-buckling delamination behavior of laminated composite with an embedded initial delamination under in-plane compression was studied experimentally and numerically. First, compression tests for laminated composite specimens with embeded initial delamination were performed and the buckling and delamination responses were obtained. Then the experimental test was numerically simulated using finite element methods with the progressive failure accounted for by using cohesive zone modeling. The load-displacement curve, strain behavior and delamination shapes of experimental specimens obtained from load cells, strain gages installed at different locations, and C scan images, respectively, were compared with the FEM results, and good agreements were attained. The effect of the buckling modes, laminate stacking sequence and shape of initial delamination on the buckling load and propagation behavior was studied by considering different ply stacking and shapes of initial delaminations. It was found that the buckling mode determined the growth direction of the delamination propagation, and the stacking sequence influenced the extent of the propagation area, while the orientation of the delamination affected the buckling loads.     

Ground surface roughness prediction based upon experimental design and neural network models

The results presented in this paper are related to the prediction of the surface roughness generated by the grinding process. The main problems associated with the prediction capability of empirical models developed using the design of experiment (DoE) method are given. The first problem is a limited aptitude to calculate an accurate minimal output value as this optimal value was found to be absurdly negative in many cases. The second problem is that these models are not able to detect particular behaviour of the outputs for particular sets of inputs. This constitutes a serious limitation of the application of this method to ground surface roughness prediction as the surface generation mechanisms differ at low and high work speed. In this study an approach suggesting the combination of DoE method and artificial neural network (ANN) is developed. x-n-1 structures using the back-propagation algorithm were selected for the developed ANNs. Data of the DoE were used to train the ANNs and the inputs of the developed ANNs were selected among the factors and the interactions between factors of the DoE depending on their significance at different confidence levels, expressed by α%. The significance was tested using the ANOVA method. Results have shown particularly, the existence of a threshold value of α% to which correspond a critical set of inputs up to which increasing the inputs, improves the learning and the prediction capability of the constructed ANNs. The built ANNs using these critical sets of inputs have shown low deviation from the training data, low deviation from the testing data and high sensibility to the inputs levels. The high prediction accuracy of the developed ANNs was conformed by the good agreement with the results of empirical models developed by previous investigations. The obtained results were valid for three kinds of steels having different properties and different hardness.     

基于递归量化观测的复合材料损伤贝叶斯估计方法

针对复合材料损伤定位定量识别中损伤指数构建困难与损伤预测的不确定性问题,提出一种基于递归量化特征观测的 损伤状态贝叶斯概率估计方法。 该方法无需基准无损信号参考与损伤波包识别,采用 Lamb 波信号的递归量化分析(RQA)特 征,从结构非线性动力学角度对复材板内部损伤进行观测,根据特征与损伤位置大小的相关性与单调性进行了敏感特征筛选。 根据各传感路径信号的 RQA 特征与损伤形态参量间的线性/ 非线性关联特性,采用多变量二次交互模型建立了损伤状态的特 征描述空间。 综合考虑各传感路径中 RQA 特征对损伤观测的不确定因素,建立了复材板损伤状态的观测模型,采用贝叶斯更 新算法对复材板损伤位置尺寸进行估计。 复材板的脱层损伤仿真分析结果表明,本文提出方法无需分析结构损伤与 Lamb 波 的复杂交互机制,就能同步实现复材板内部损伤的量化评估与定位分析,估计出的损伤位置大小在仿真模型的损伤参数设定值 的 75% 置信区间内,损伤估计成像区域与仿真损伤区域相符合。 复合材料板的疲劳损伤试验分析结果表明,采用本文方法经过 24 次迭代更新后,得到的损伤面积与大小完全覆盖了 X 射线扫描检测的真实损伤。 该方法充分考虑了复材板损伤评估中的不 确定因素,更加贴合复合材料结构损伤检测与评估的工程实际应用,具有较好的应用前景。     

A simple method to calculate mode components of strain energy release rate of free-edge delaminations in composite laminates

A simple method, which calculates the mode components of the strain energy release rate of free-edge delaminations in the laminates, is proposed. The interlaminar stresses are evaluated as an interface moment and interface shear forces that are obtained from the equilibrium equations at the interface between the adjacent layers. Deformation of an edge-delaminated laminate is calculated by using a generalized quasi-three dimensional classical lamination theory developed by the author. The analysis provides closed-form expressions for the three components of the strain energy release rale. The analyses are performed on [+30/−30/90]s laminates subjected to uniaxial extension, with free-edge delaminations located symmetrically and asymmetrically with respect to the laminate midplane. Comparison of the results with a finite element solution using the virtual crack closure technique shows good agreement. The simple nature of this method makes it suitable for primary design analysis for the delaminations of composite laminates.     

神经网络在逆向工程中进行三维实体特征识别

介绍了一种基于神经网络直接从测量数据中提取三维实体特征的方法，包括以下几个部分：（1）点云边点识别和区域分割；（2）对点云边点进行拟合，形成特征边，并进行特征编码；采用ANN进行特征识别；（3）由边特征信息提取特征参数，构造三维实体特征。由神经网络方法直接提取实体特征使逆向工程集成制造系统的建立更为方便。     

Asymptotic approach to the constant velocity of hydrogen delamination growth

We consider delamination crack growth controlled by gas diffusion into the crack. If the gas is accumulated inside the delamination, after some incubation period, the crack starts growing under the pressure of the accumulated gas. An important example is given by hydrogen-induced delamination. Hydrogen absorbed by a metal is typically dissolved in the proton form within the lattice. Some of the protons reach the surface of pre-existing or freshly created cracks or delaminations where they recombine with electrons and form molecular hydrogen in the crack cavity. Since the molecular form of hydrogen is thermodynamically more stable, this process leads to accumulation of hydrogen gas inside the delamination crack. Under the excessive hydrogen pressure fracture often takes place even in the absence of any additional external loading. It is especially dangerous if the metal is covered with coating, where because of the dissimilarity of materials, microscopic voids appear more frequently. This leads first to hydrogen precipitation within these coating–metal interface voids and then to their development. This results in the delaminating of the coating. As it is well known, in the absence of aggressive media the cracking resistance of a material is characterized by a unique constant: so that if the stress intensity factor is smaller than this constant, the crack is fixed, and if it is larger, the crack moves in a dynamic regime. But in the presence of aggressive media, particularly hydrogen, the crack development is characterized by a smooth kinetic function v(K_I), which is dependence of the crack velocity on the stress intensity factor, with the lower threshold value of resistance K_scc smaller than the static critical stress intensity factor K_Ic. As numerically shown in the author's earlier work for the internal crack growth, the crack velocity first increases in accordance with the kinetic function, until reaching some value of v_s, and remains at the same level afterwards. Since, as has been obtained by the author in the previous paper, kinetic equations for internal and delamination cracks are essentially identical, the same conclusion can be derived for the latter. In this paper the stability and asymptotic approach to the constant velocity of delamination growth is proved analytically. If v_s is known, the corresponding value of the stress intensity factor, K_s, is obtained by substituting v_s into the kinetic function v(K_I).     

Vortex flowmeter with enhanced turndown ratio based on high-frequency pressure sensors and improved convection velocity estimation

A novel vortex flowmeter by using pressure sensors and improved convection velocity estimation was proposed to extend the lower operation range of flowrate. The two sensors were mounted on the positions of 0.2D45° and 1.0D45° according to the signal strength and vortex wavelength criterion. The high-frequency pressure sensors were designed to acquire the undistorted signals of vortex-induced pressure fluctuation and further compute the vortex convection velocity. Aiming at the problem of multiple peaks in cross-correlation calculation, a modified transit time estimation technique combined with a moving-average-filter-based cross correlation function were introduced and verified by the tests. Then, the mean convection coefficient was obtained and the performance of the transit-time-based method was analyzed and compared with that of the frequency-based method. It indicated that the novel vortex flowmeter is robust at low Reynolds number range, which achieves a turndown ratio of 8:1 with an accuracy of ±2% in the Reynolds number range from 1.53×10⁴ to 1.23×10⁵. A remarkable improvement of turndown ratio is achieved compared to the original frequency-based method of 3:1. The proposed system of pressure sensors shows good prospect for the gas metering in small pipes due to the feature of non-invasion and sufficient high frequency response.     

基于ANN/GA的汽车覆盖拉延件要领设计专家系统

提出了要领设计的新概念， 即通过该环节为汽车覆盖拉延件成形的有限元模拟提供初始条件， 分析了要领设计的必要性和特点。提出将流入量 （从压料面下进入凹模型腔的局部材料线素长度） 作为衡量局部成形阻力的指标。通过人工神经网络由覆盖件侧壁高度得到局部流入量， 将其添加在拉延件轮廓 （在成品件上添加工艺补充得到） 外侧， 即得初始毛料外形。把拉延筋的设计原则转化为约束条件， 将遗传算法和神经网络用于拉筋及凹模圆角几何参数的优化设计。另外还阐述了特征建模、面向对象知识库的应用。     

Examination and modification of equivalent delamination factor for assessment of high speed drilling

Composite components suffer delamination at the entrance and exit of drilled holes. Many measures have been suggested by different researchers to assess such delamination damage. These include delamination factor, two-dimensional delamination factor, damage ratio, adjusted delamination factor, refined delamination factor, equivalent delamination factor, and minimum delamination factor. Among all these various assessment factors, the equivalent delamination factor looks simple and able to take into account the different features of delamination. However, the method of calculation of the equivalent delamination factor may not provide accurate values for delamination resulting from high speed drilling. The aim of this paper is to evaluate the equivalent delamination factor in high speed drilling of a composite laminate using a twist drill and develop a new approach to determine equivalent delamination factor which can be used for both conventional and high speed drilling conditions. This new method is applied to calculate the equivalent delamination factor in trials of drilling composite specimens at different speeds and feed rates and is found suitable.

     

基于GMKL-SVM的模拟电路故障诊断方法

提出了一种新颖的基于广义多核支持向量机(GMKL-SVM)的模拟电路故障诊断方法。首先,应用Haar小波分析提取被测电路时域响应信号的小波系数作为特征参量,并生成样本数据;然后,基于样本数据,应用量子粒子群算法对GMKL-SVM的参数进行优化,并以此建立基于GMKL-SVM的故障诊断模型,用于区分模拟电路的各个故障。实例电路的单故障和双故障诊断实验结果表明,所提出的GMKL-SVM方法能较好地实现模拟电路故障诊断,与传统的GMKL-SVM方法相比,表现出了更好的性能,获得了更高的故障诊断正确率。     

A novel approach of fractional-order time delay system modeling based on Haar wavelet

In this paper, fractional-order time delay system modeling is presented using Haar wavelet operational matrix of integration. Therefore, it does not require any prior knowledge of transfer function structure or partial information about fractional differentiation order. It allows the estimation of the implicit time delay parameter together with other model parameters by utilizing new delay operational matrix of Haar wavelet based on Riemann-Liouville definition. The proposed technique reduces the complexity of identification by converting the complex fractional calculus equations into simple algebra. The efficacy of the approach is verified on various integer and non-integer (fractional) order systems in simulation. For realistic condition, proposed method is verified in the presence of noise in simulation and also demonstrated on the real-time process control temperature system.     

受载复合材料曲板结构的失效特征研究

为研究飞机复合材料曲板的失效过程与失效模式,对一组复合材料弯角试验件进行了准静态拉伸试验.同时基于ABAQUS软件及其USDFLD子程序,考虑复合材料的渐进损伤特征,建立了受载曲板的有限元模型,分析弯角区的应力分布特征.研究发现：数值仿真能够较准确的模拟弯角结构的受载过程,模拟结果与试验结果吻合较好;结构受载过程中,在弯角区域发生分层破坏,随着载荷的增大,分层沿着周向和径向都有扩展;随着外加载荷的增大,弯角区域的最大法向应力的位置逐渐移向加载端.