Analysis of Laminated Rubber Bearings with a Numerical Reduction Model Method

In this paper, we present a reduction method for modeling slender laminated elastomeric structures, which is developed in the context of nearly incompressible hyperelasticity. This method, based on a finite element formulation, consists in projecting the unknown fields onto a polynomial basis in order to reduce the dimension of the problem and the model size. Two types of finite elements are used, one for plane-strain and the other for 3D structures. Comparisons with classical finite element models on single layers show the reliability of the present method. The method proposed successfully predicts the global and local behavior and since it reduces both the model size and the computing time. It can be used to model slender bearing consisting of several layers.     

High-order time integration applied to metal powder plasticity

The present article treats two objectives. In the first investigation attention is focused on the application of time-adaptive finite elements formulated on the basis of a high-order time integration procedure on a constitutive model for compressible finite strain viscoplasticity for metal powder. In this connection, it has to be emphasized that the integration procedure is not only applied to the evolution equations on Gauss-point level but on the total system of differential–algebraic equations resulting from the application of the vertical line method on the quasi-static finite element equations. The specific application emerges from the field of metal powder compaction. Particular studies are carried out using stiffly accurate, diagonally implicit Runge–Kutta methods in combination with the Multilevel-Newton algorithm for solving the DAE-system. In this respect, the effort vs. accuracy behavior is investigated which is also related to order reduction known in elastoplasticity. The second topic treats the local stress algorithm for taking into account the yield function based finite strain viscoplasticity model, where the classical Newton–Raphson method fails. This is the reason why most constitutive models of powder materials are implemented into explicit finite element codes. Thus, the proposed investigations compare different methods in view of a stable and efficient integration process in implicit finite element formulations.     

基于局部插值的结构动力模型降阶方法

提出了一种基于局部插值对大型结构有限元模型的特征值问题进行降阶的方法. 该方法通过局部插值将复杂结构的有限元模型中节点的位移用凝聚点的位移插值来表示, 从而得到用插值函数表示的简化基向量, 实现对结构广义特征值问题的降阶. 为了提高降阶模型的精度, 采用非协调元的插值函数作为局部插值函数来弱化凝聚后的结构刚度, 并且在有限元模型上进行逆迭代, 对得到的降阶后的广义特征值问题的特征值和特征向量进行改善. 为了提高模型降阶的效率, 采用规整网格包围整个结构生成均匀的凝聚点, 高效地确定了有限元模型中节点所依附的凝聚点. 最后, 对 3 个机床部件的模态分析验证了提出的简化方法的高效性和可行性.      

考虑粘结面厚度的局部粘结单元法

传统无厚度粘结单元法CFEM (Cohesive finite element method)在模拟脆性材料断裂方面具有很强的优势,但也存在很大问题.一是单元尺寸增大,收敛性变差;二是单元尺寸变小,模型刚度发生折减.为了克服这两个问题,发展了考虑厚度的局部粘结单元法,即在裂纹可能扩展区插入具有一定厚度的粘结面单元.粘结面单元采用拓展虚内键本构(Augmented virtual internal bond)描述.由于考虑了厚度,粘结面交叉处会形成多边形空缺.为了弥补这一空缺,将其看作多边形键元胞,采用离散虚内键模型(Discretized virtual internal bond)对其建模,保证了模型的几何完整性.模拟结果表明,本文方法有效,克服了传统CFEM方法的刚度折减问题,提高了计算稳定性和收敛性.     

Identification of dynamic properties of plate-like structures by using a continuum model

The common approach currently used in the aircraft structural analysis is the finite element method. NASA's research in computational structures technology (CST) is helping to develop the finite element analysis to a new stage, although the significant limitations still exist. The elements used in the finite element method are usually void of dynamics. The consequence is that hundreds and thousands of elements are needed to represent large flexible aircraft structures in order to acquire analytical accuracy. To avoid the large dimensionality the current practice is to reduce the order of the model for structural system identification and control synthesis. This approximation, however, can lead to system instability due to the dynamics which are ignored.In contrast, distributed parameter modeling seems to offer a viable alternative to the finite element approach for modeling large flexible aerospace structures. Distributed parameter models have the advantage of improved accuracy, reduced number of modal parameters, and the avoidance of modal order reduction. Most of the effort on the continuum modeling so far is contributed to the beam-like structures which are composed of beams, tethers and rigid bodies. For the aircraft structural analysis, however, another important type of structural elements is plate. The principle of the monocoque or semi-monocoque type of aircraft construction is fundamentally the use of a thin-walled tube to carry compression, tension, shear, and bending. It is necessary, therefore, to expand the continuum modeling methodology to the plate-like structures to satisfy the requirement in the aircraft structural analysis, especially for the monocoque structures.This paper has developed a continuum modeling algorithm for the identification of dynamic properties of plate-like structures. A closed-form solution of the Timoshenko plate equation consistent with the maximum likelihood estimator has been derived. The closed-form expressions of the gradient functions have thereby been resulted from the solution of the partial differential equation. The proposed distributed parameter model involves far fewer unknown parameters than independent modal characteristics for finite element models. Illustration of this approach is given by a computer simulation which shows that the estimated results by using continuum model are reasonably accurate compared with the theoretical results.     

Adaptive shape functions and internal mesh adaptation for modeling progressive failure in adhesively bonded joints

Macroscopic finite elements are elements with an embedded analytical solution that can capture detailed local fields, enabling more efficient, mesh independent finite element analysis. The shape functions are determined based on the analytical model rather than prescribed. This method was applied to adhesively bonded joints to model joint behavior with one element through the thickness. This study demonstrates two methods of maintaining the fidelity of such elements during adhesive non-linearity and cracking without increasing the mesh needed for an accurate solution. The first method uses adaptive shape functions, where the shape functions are recalculated at each load step based on the softening of the adhesive. The second method is internal mesh adaption, where cracking of the adhesive within an element is captured by further discretizing the element internally to represent the partially cracked geometry. By keeping mesh adaptations within an element, a finer mesh can be used during the analysis without affecting the global finite element model mesh. Examples are shown which highlight when each method is most effective in reducing the number of elements needed to capture adhesive nonlinearity and cracking. These methods are validated against analogous finite element models utilizing cohesive zone elements.     

主动约束层阻尼梁的数值模型

为对主动约束层阻尼结构建立精确完善的数学模型，采用有限元建模，并考虑到压电材料的机电耦合效应和粘弹性材料的本构关系随温度、频率的变化而变化的特点，将有限元方法与粘弹性材料的GHA模型相结合，从而避免因粘弹性材料导致的非线性微分方程，能直接求解模态频率、模态阻尼及结构响应。为进一步设计控制器，先在物理空间进行动力缩聚，将系统降至适当的维数，然后在状态空间用鲁棒防阶的方法进一步降阶。这样既能大大降低系统维数，又能保证降阶后系统稳定、可控、可观。这对于重量轻、柔度大、低频密集的大型空间柔性结构尤其重要。     

Nonlinear finite element analysis of inflatable beams made from orthotropic woven fabric

This paper was devoted to the three-dimensional nonlinear finite element analysis of inflatable beams. The beams under consideration are made of modern textile materials and can be used as a load-bearing beams or arches when inflated. A 3D Timoshenko beam with a homogeneous orthotropic woven fabric (OWF) was proposed. The model took into account the geometric nonlinearities and the follower force resulting from the inflation pressure. The use was made of the usual total Lagrangian form of the virtual work principle to perform the nonlinear equilibrium equations which were discretized by the finite element method. Two kinds of solutions were then investigated: finite elements solutions for linearized problems which were obtained by the means of the linearization around the prestressed reference configuration of the nonlinear equations and nonlinear finite element solutions which were performed by the use of an optimization algorithm based on the Quasi-Newton method. As an example, the bending problem of a cantilever inflated beam under concentrated load was considered and the deflection results improve the existing theoretical models. As these beams are made from fabric, the beam models were validated through their comparison with a 3D thin-shell finite element model. The influence of the material effective properties and the inflation pressure on the beam response was also investigated through a parametric study. The finite elements solutions for linearized problems were found to be close to the theoretical results existing in the literature. On the other hand, the results for the nonlinear finite element model were shown to be close to the results for the linearized finite elements model in the case of high mechanical properties and the nonlinear finite element model was used to improve the linearized model when the mechanical properties of the fabric are low.     

基于机器学习软骨细胞的时间依赖性力学行为及本构参数反演

探究软骨细胞机械负载下的力学特性对于理解软骨细胞的正常和病理状态以及骨性关节炎的病因至关重要. 基于软骨细胞有限元计算模型的力学响应与其本构参数之间的高度复杂非线性, 本文提出了分别利用双向深度神经网络TW-Deepnets模型和随机森林RF模型并结合有限元方法来识别软骨细胞本构参数的两种反演方法. 首先, 建立了软骨细胞的无侧限压缩实验有限元模型, 收集MSnHS本构参数空间点与对应的有限元计算模型的压缩反作用力响应数据集. 其次, 结合贝叶斯超参数优化算法搭建了用于软骨细胞本构参数反求的TW-Deepnets模型和RF模型, 对有限元收集的数据进行训练, 并利用单个软骨细胞受到50%压缩程度下的实验数据对软骨细胞的MSnHS本构参数进行了反求. 最后, 通过与实验曲线的对比验证了所提出的反演方法的有效性, 并引入决定系数R2对两种模型的预测准确性进行了对比评估, 检验了模型对各本构参数的预测性能, 分析了MSnHS本构模型中各参数影响软骨细胞力学响应的重要性占比. 结果表明, 本研究提出的本构参数反演方法能够有效获取软骨细胞的本构参数值, 从而准确描述软骨细胞的时间依赖性力学特性, 该方法也可进一步推广到生物细胞在静态或动态负载条件下的复杂参数反演问题.      

Reduced order model of offshore wind turbine wake by proper orthogonal decomposition

A wide range of previously designed methods for faster parametrization of partial differential equations requires them to be solved using existing finite volume, finite element, and finite difference solvers. Due to the requirement of high degrees of freedom to accurately model the physical system, computational costs often becomes a bottle-neck. It poses challenges to conducting efficient repeated parametric sampling of the input parameter that disrupts the whole design process. Model reduction techniques adopted to high fidelity systems provide a basis to accurately represent a physical system with a lower degree of freedom. The present work focuses on one such method for high-fidelity simulations that combines finite volume strategy with proper orthogonal decomposition and Galerkin projection to test reduced-order models for high Reynolds number flow applications. The model is first benchmarked against flow around a cylinder for which extensive numerical and experimental data is available in the literature. The models are then tested to full-scale NREL 5MW offshore wind turbines to evaluate wake evolution in the downstream direction. The simulations results show relative errors of wind turbines for the first seventy modes approach 4.7% in L²-norm for velocities.     

A comparison of projection-based model reduction concepts in the context of nonlinear biomechanics

Computational assistance gains increasing importance in the field of medical surgery. As an example, in the present work, we look at functional endoscopic sinus surgery. Simulations for surgery training programs or online support during surgeries require simulation tools which are characterized by a preferably short simulation time (real time) and a high degree of accuracy. The nonlinear finite element method is most suitable to yield qualitatively and quantitatively reliable results. The problem is, however, to achieve such results in real time. One possibility to reach both, short computational time and high accuracy, is to combine model reduction and finite element techniques. Therefore, in this paper, various projection-based model reduction methods are discussed and compared with respect to their possible application in biomechanics. The modal basis, the load-dependent Ritz and the proper orthogonal decomposition (POD) method were used to reduce the model of a cube under compression considering different material nonlinearities and large deformations. The POD method led to the lowest errors in displacement and stress while providing the largest reduction in CPU time. Further, the influence of different POD parameters was investigated. According to this study, the snapshots upon which the POD is based had to agree as closely as possible with the original deformation of the reduced system. The POD method applied to the finite element model of an inferior turbinate led to an adequate accuracy for surgery simulations within less than one-third of the computational time of the unreduced finite element simulation.     

Dynamic buckling analysis of composite cylindrical shells using a finite element based perturbation method

In this paper a finite element formulation of a reduction method for dynamic buckling analysis of imperfection-sensitive shell structures is presented. The reduction method makes use of a perturbation approach, initially developed for static buckling and later extended to dynamic buckling analysis. The implementation of a single-mode dynamic buckling analysis in a general purpose finite element code is described. The effectiveness of the approach is illustrated by application to the dynamic buckling of composite cylindrical shells under axial and radial step loads. Results of the reduction method are compared with results available in the literature. The results are also compared with full model finite element explicit dynamic analysis, and a reasonable agreement is obtained.     

Multiscale modelling of the plastic anisotropy and deformation texture of polycrystalline materials

A hierarchical multilevel method is presented for the plastic deformation of polycrystalline materials with texture-induced anisotropy. It is intended as a constitutive material model for finite element codes for the simulation of metal forming processes or for the prediction of forming limits. It consists of macroscopic models of which the parameters are to be identified using the results of two-level (meso/macro) or three-level (micro/meso/macro) models. A few such two-level models are presented, ranging from the full-constraints Taylor model to the crystal-plasticity finite element models, including the grain interaction models GIA, LAMEL and ALAMEL. Validation efforts based on experimental cold rolling textures obtained for steel and aluminium alloys are shortly discussed. An assessment is also given of the assumptions of the LAMEL and ALAMEL models concerning stress and strain rate heterogeneity at grain boundaries, based on the results of a crystal plasticity finite element study. Finally a recent three-level model which also looks at the microscopic level (dislocation substructure) is discussed.     

Analysis of thermo-mechanical behaviour of a crack using XFEM for Leak-before-Break assessments

The stresses near a crack which has a fluid escaping through it are presented in this paper. The pressure and heat flux, due to the fluid acting on the crack walls, are imposed as boundary conditions in a new finite element tool which has been developed specifically for Leak-before-Break. This special tool uses the extended finite element method to include information about the problem on a sub element level. It is shown to be as accurate as standard finite element models which use very refined meshes, but having the added benefit of being much quicker to implement, and vastly reducing postprocessing. This means that leak rates can be investigated more efficiently. The model is thermo-elastic, and plasticity is accounted for by a correction to the crack opening displacement based on the R6 method. Both crack opening area and peak stresses are shown to decrease when the walls of the crack are hotter than the background plate temperature. The consequences of this for Leak-before-Break assessments are discussed in the paper.     

Continuum damage models with non-conventional finite element formulations

In recent years, some research effort has been devoted to the development of non-conventional finite element models for the analysis of concrete structures. These models use continuum damage mechanics to represent the physically non-linear behavior of this quasi-brittle material. Two alternative approaches proved to be robust and computationally competitive when compared with the classical displacement finite element implementations. The first corresponds to the hybrid-mixed stress model where both the effective stress and the displacement fields are independently modeled in the domain of each finite element and the displacements are approximated along the static boundary, which is considered to include the inter-element edges. The second approach corresponds to a hybrid-displacement model. In this case, the displacements in the domain of each element and the tractions along the kinematic boundary are independently approximated. Since it is a displacement model, the inter-element boundaries are now included in the kinematic boundary. In both models, complete sets of orthonormal Legendre polynomials are used to define all approximations required, so very effective p-refinement procedures can be implemented. This paper illustrates the numerical performance of these two alternative approaches and compares their efficiency and accuracy with the classical finite element models. For this purpose, a set of numerical tests is presented and discussed.     

单变量有限元的新思考:精化直接刚度法

对单变量有限元和多变量有限元进行了综合分析和比较。其中包括有限元法的单元的可靠性问题:收敛性,坐标不变性,伪零能模式;单元的性能:计算精度,对畸变网格的适应性及解除闭锁、解的晃动现象等问题。进一步对单变量有限元提出新的思考,用精化直接刚度法发展单变量有限元,使之能兼备单变量有限元与多变量有限元之长,克服两者之短。     

Finite element modeling and optimization to prevent wire breakage in electro-discharge machining

One of the most important problems in wire electrical discharge machining is related to wire breakage. This research develops a simple finite element model and a new approach to predict the thermal distribution in the wire fairly accurately. The model can be used to optimize the different parameters of the system to prevent wire breakage. At any instant of time, the spatial heat distribution profile of the wire can be mapped on the transient analysis of any point on the wire traversing through all the heat zones from the top spool to the bottom end. Based on this principle, the finite element model and optimization algorithm are used to determine that the heat generated is the critical variable responsible for wire breakage. The model successfully predicts the thermal distribution profile accurately for various wire materials, for increased wire velocity and for reduction in heat transfer coefficient. This simple model is a precursor of development for 3-D finite element models that can describe the cross-sectional wire erosion as the workpiece cutting progresses. The modeling may lead to the development of a smart electro-discharge machining system with a sensor and feedback control to increase the cutting speed and minimize breakage.     

求解固体力学问题的强?弱耦合形式单元微分法

单元微分法是一种新型强形式有限单元法. 与弱形式算法相比, 该算法直接对控制方程进行离散, 不需要用到数值积分. 因此该算法有较简单的形式, 并且其在计算系数矩阵时具有极高的效率. 但作为一种强形式算法, 单元微分法往往需要较多网格或者更高阶单元才能达到满意的计算精度. 与此同时, 对于一些包含奇异点的模型, 如在多材料界面、间断边界条件、裂纹尖端等处, 传统单元微分法往往得不到较精确的计算结果. 为了克服这些缺点, 本文提出了将伽辽金有限元法与单元微分法相结合的强?弱耦合算法, 即整体模型采用单元微分法的同时, 在奇异点附近或某些关键部件采用有限元法. 该策略在保留单元微分法高效率与简洁形式等优点的同时, 确保了求解奇异问题的精度. 在处理大规模问题时, 针对关键部件采用有限元法, 其他部件采用单元微分法, 可以在得到较精确结果的同时, 极大提高整体计算效率. 在本文中, 给出了两个典型算例, 一个是具有切口的二维问题, 一个是复杂的三维发动机问题. 针对这两个问题, 分析了该耦合算法在求二维奇异问题和三维大规模问题时的精度与效率.      

基于卷积非粘滞阻尼模型的超高层结构风效应分析

有限元方法中相对于对结构质量与刚度特性的描述,结构阻尼的描述仍具有较大的模糊性。随着新型建筑材料与复杂结构体系的发展,以及对计算机模拟要求的提高,阻尼作用的机理与相应阻尼模型的研究成为值得关注的问题。基于一种阻尼力与质点速度历程相关的卷积非粘滞阻尼模型,采用微分求积求解算法,对一个大型复杂超高层建筑结构的风振响应进行了分析,并与常用的比例粘滞阻尼模型进行了对比。对卷积非粘滞阻尼力模型系统的响应特征进行了分析,特别是该模型的松弛效应对结构响应的影响。另外,作为将这种新阻尼模型应用于实际工程的一次探索,本文采用微分求积算法,建立了一套可将该阻尼模型及其求解算法嵌入通用有限元软件的求解系统,可用于复杂结构的动力响应分析。     

Réduction a priori de modèles thermomécaniquesAn a priori model reduction method for thermomechanical problems

A model reduction method is proposed for finite element models. A previous computation of the state of the structure is not necessary. Residuals defined over the entire time interval and the Karhunen–Loève method provide basis functions. A non-incremental algorithm, from the LATIN method, is used to compute this basis functions. Because of the non-incremental feature, the reduced order model is representative for a large evolution of the state of the structure. To cite this article: D. Ryckelynck, C. R. Mecanique 330 (2002) 499–505.