基于分段Knothe函数的采空区埋地管道动态力学特性研究

针对穿越采空区埋地管道的动态力学预测问题,本文在概率积分法基础上将煤矿开采距离定义为开采时间与开采速度的乘积,采用分段Knothe函数模型并使用叠加原理,建立了在管-土协同变形期间水平煤层及缓倾斜煤层下埋地管道的动态下沉模型;在此基础上,运用弹性地基梁模型求解管道的挠度并结合分段Knothe函数建立了管道动态力学预测模型。结果表明:通过对比理论计算数据与实测数据可知,基于分段Knothe函数的预测方法能够比较真实地反映埋地管道的动态下沉位移及不断开采对埋地管道下沉变化的客观规律,且通过管道挠度方程能准确预测管道的动态力学行为。     

非局部应变梯度理论下纳米梁的力学特性研究

基于非局部应变梯度理论,建立了一种具有尺度效应的高阶剪切变形纳米梁的力学模型.其中,考虑了应变场和一阶应变梯度场下的非局部效应.采用哈密顿原理推导了纳米梁的控制方程和边界条件,并给出了简支边界条件下静弯曲、自由振动和线性屈曲问题的纳维级数解.数值结果表明,非局部效应对梁的刚度产生软化作用,应变梯度效应对纳米梁的刚度产生硬化作用,梁的刚度整体呈现软化还是硬化效应依赖于非局部参数与材料特征尺度的比值.梁的厚度与材料特征尺度越接近,非局部应变梯度理论与经典弹性理论所预测结果之间的差异越显著.     

关于弹性梁的数学模型

叙述和比较一维弹性体的两种不同建模方法, 即弹性梁的传统建模方法和基于 Kirchhoff-Cosserat模型的建模方法. 应用精确Cosserat模型分析梁的三维运动. 考虑中 心线的拉伸压缩变形、截面的剪切变形、截面转动的惯性和端部载荷影响等因素, 建立精确 的弹性梁动力学方程. 讨论梁的静态和动态平衡稳定性. Kirchhoff杆、铁摩辛柯 梁和欧拉--伯努利梁等为Cosserat模型在各种简化条件下的特例.     

平面弹性力学问题的离散元法

根据离散元的基本原理,基于变形体的理论提出了适用于平面弹性力学问题的界面位移、应变和应力模式,建立了求解平面弹性力学问题的离散元方程和相应的迭代求解方法.通过界面位移可以简洁地将位移和力的边界条件引入离散系统的控制方程,也可以方便地求解节点位移.数值算例表明,与具有相同网格的有限元结果相比,离散元能同时给出精度相对较高的应力解和精度相当的位移解.     

超细长弹性杆的分析力学问题

超细长弹性杆作为DNA等生物大分子链的力学模型，其平衡和稳定性问题已成为力学与分子生物学交叉的研究热点．虽然在Kirchhoff动力学比拟的基础上，用分析力学方法讨论弹性杆的文章已见诸文献，但尚未形成弹性杆分析力学的严格理论．本文研究了超细长弹性杆分析力学的若干基础性问题．对杆截面的自由度、虚位移、约束方程及约束力等基本概念给出严格的定义和表达式．建立弹性杆平衡的D’Alembert-Lagrange原理、Jourdain原理和Gauss原理；从D’Alembert-Lagrange原理导出Hamilton原理．从变分原理出发导出Lagrange方程、Nielsen方程、Appell方程和Hamilton正则方程；对于受约束的弹性杆，导出了带乘子的Lagrange方程．讨论了Lagrange方程的首次积分．对于杆中心线存在尖点的情形，导出了微段杆平衡的近似方程。     

分析力学初值问题的一种变分原理形式

明确了分析力学初值问题的控制方程，按照广义力和广义位移之间的对应关系，将 各控制方程卷乘上相应的虚量，代数相加，进而在 原空间中建立了分析力学初值问题的一种变分原理形式，即建立了分析力学初值问题的卷积 型变分原理和卷积型广义变分原理. 推导了分析力学初值问题卷积型变分原理和卷积型广义 变分原理的驻值条件. 在建立分析力学初值问题的一种变分原理形式的同时, 将变积方法推广为卷变积方法.     

考虑尺度效应的微梁静态弯曲数值分析

基于修正偶应力理论及表面弹性理论,本文提出了一种新的双曲线剪切变形梁模型,用于均匀微尺度梁的静态弯曲分析。该理论可以直接利用本构关系获得横向剪切应力,满足梁顶部和底部的无应力边界条件,避免了引入剪切修正因子。根据广义Young-Laplace方程建立了梁的内部与表面层的应力连续性条件,单一的变量场可以描述梁的位移模式。通过在位移场中考虑表面层厚度以及表面层的应力连续条件,可以使新模型能够更准确地预测微尺寸和表面能相关的尺度效应。通过Hamilton原理推导出了梁的控制方程和边界条件。应变能除了考虑经典弹性理论,还要考虑微结构效应和表面能。Navier-type的解析解适用于简支边界条件,而基于拉格朗日插值的微分求积法(DQEM)可以研究在不同边界条件下的力学响应。把该数值解与Navier方法得出的解析解作了对比,得出:微尺度梁在考虑表面能或微尺寸效应、不同载荷和梁高变化下的响应一致;当不考虑微结构相关性和表面能效应时,该模型退化为经典的欧拉梁模型。     

基于等效阻尼理论的金属橡胶弹性迟滞力学模型及实验研究

针对弹性多孔金属橡胶非线性迟滞特性力学行为,将迟滞恢复力-位移曲线分解为非线性单值曲线和椭圆,并将等效阻尼理论用于动态力学性能参数识别,从而建立了一种新型的适用于黏弹性阻尼材料的宏观唯象力学模型。采用不同相对密度的环形金属橡胶进行动态实验测试,以验证理论模型的准确性,结果表明该模型可将具有非线性特性的金属橡胶系统进行降阶处理,提高金属橡胶力学模型的预测效率,并能很好地描述金属橡胶的迟滞力学行为。另外,研究了在不同激励频率条件下金属橡胶的阻尼耗能特性。实验结果表明:在高频加载的条件下,黏性阻尼系数对动态加载频率不敏感,阻尼耗能与加载幅值之间呈线性正相关。基于等效阻尼理论的弹性迟滞力学模型具有一定的普适性,可进一步推广应用于类似弹性多孔材料的力学性能表征,为其工程应用提供理论基础。     

热弹性力学的广义变分原理

Biot建立了热弹性力学的变分原理。此后,和Ⅲ将上述变分原理推广到有热源的情况,从而导出了热弹性力学的力学平衡方程,力的边界条件以及具有热源的热传导方程。 下面建立带有运算子的热弹性力学的广义变分原理。根据此原理可以导出力学平衡     

有限到无限:基于离散静力平衡推导悬链线方程和梁位移方程

悬链线问题是一类非常经典的力学问题,推导出悬链线方程的方法从静力平衡到变分原理,非常之多。本文基于离散的静力平衡模型,采用从有限到无限的求解方法,建立并求解了悬链线方程。在此基础上,本文进一步建立了含有扭簧作用的离散静力平衡模型,推导并求解梁位移方程。最后在不同参数下求解梁位移方程得到其渐变解,并分析了多种平衡状态的存在情况。     

实验数据反演识别分析方法与力学表征

本文给出了两类基于实验(实测)数据的反演识别方法,简述了其在界面力学性能及工程实测数据分析中的应用.在粘接界面力学性能的反演识别实验研究中,基于不同加载速率的实验曲线,结合参数化界面力学模型,通过反演识别给出粘接界面力学性能参数,并对识别结果的适定性进行独立的实验验证;在基于工程实测数据的反演识别与力学建模方面,在分析盾构装备载荷特点的基础上,对海量的实测数据统计分类,提出了力学量纲分析的建模方法,并应用于盾构掘进载荷的反演识别研究.     

基于深度强化学习算法的颗粒材料应力?应变关系数据驱动模拟研究

颗粒材料的宏观力学行为受颗粒组分等材料参数, 孔隙率、配位数等状态参数的影响, 同时又具备复杂的加载路径和加载历史相关性, 建立包含多个内变量以及各变量间相互关联的颗粒材料本构模型是一个重要的科学难题. 不同于传统的基于屈服面、流动法则和硬化函数框架下的唯象本构模型, 本文基于颗粒物质力学的研究基础, 以颗粒材料平均孔隙率、细观组构参数和弹性刚度参数作为内变量, 结合深度学习方法建立以有向图表征的数据本构模型. 有向图中以不同的链接网络表示不同的内变量信息流动方向, 各个内变量间的映射关系采用循环神经网络来建立, 将各个神经网络相互组合, 形成包含不同内变量且具有不同预测能力的本构模型. 该本构模型的建立过程等价于在众多可能的内变量链接关系空间中寻找最能描述实际材料宏观应力应变行为的优化问题. 因此, 可将有向图本构模型的建立过程看作“马尔可夫决策过程”, 采用深度强化学习算法构建有向图的内变量链接组合优化过程, 具体采用AlphaGo Zero算法自动寻找最优的颗粒材料数据驱动本构模型建模路径. 研究结果表明, 采用有向图和深度强化学习算法可建立起完全依靠“数据驱动”的颗粒材料应力?应变关系. 此外, 本方法提供了一种将不同理论模型从数据角度统一起来, 且基于人工智能算法发展更优模型的研究思路, 可为相似问题的研究提供借鉴.      

Nonlinear dynamics of three-dimensional belt drives using the finite-element method

In this paper, new nonlinear dynamic formulations for belt drives based on the three-dimensional absolute nodal coordinate formulation are developed. Two large deformation three-dimensional finite elements are used to develop two different belt-drive models that have different numbers of degrees of freedom and different modes of deformation. Both three-dimensional finite elements are based on a nonlinear elasticity theory that accounts for geometric nonlinearities due to large deformation and rotations. The first element is a thin-plate element that is based on the Kirchhoff plate assumptions and captures both membrane and bending stiffness effects. The other three-dimensional element used in this investigation is a cable element obtained from a more general three-dimensional beam element by eliminating degrees of freedom which are not significant in some cable and belt applications. Both finite elements used in this investigation allow for systematic inclusion or exclusion of the bending stiffness, thereby enabling systematic examination of the effect of bending on the nonlinear dynamics of belt drives. The finite-element formulations developed in this paper are implemented in a general purpose three-dimensional flexible multibody algorithm that allows for developing more detailed models of mechanical systems that include belt drives subject to general loading conditions, nonlinear algebraic constraints, and arbitrary large displacements. The use of the formulations developed in this investigation is demonstrated using two-roller belt-drive system. The results obtained using the two finite-element formulations are compared and the convergence of the two finite-element solutions is examined.     

Continuum modeling of periodic brickwork

Continuum models of periodic masonry brickwork, viewed at a micro-level as a discrete system, are identified within the frame of linearized elasticity. The accuracy of various identification schemes is investigated for standard and micropolar continua, which are directly compared with the help of some numerical benchmarks, for different loading conditions that induce periodic and non-periodic deformation states. It is shown that periodic deformation states of brickwork are exactly reproduced by both continua, provided that a suitable identification scheme is adopted. For non-periodic states micropolar continuum is shown to better reproduce the discrete solutions, due to its capability to take scale effects into account. Both continua are asymptotically equivalent as the characteristic length of the discrete system tends to zero, while providing an upper and a lower bound of the discrete solution.     

Enabling reduced-order data-driven nonlinear identification and modeling through naïve elastic net regularization

This work discusses an improved method of reduced-order modeling for existing data-driven nonlinear identification techniques through the incorporation of naïve elastic net regularization. The data-driven methods considered for this study operate using basis functions to represent the observed nonlinearity. Elastic net regularization is used to minimize the number of non-zero coefficients, thus modifying the basis functions and providing a compact representation. The ability of the naïve elastic net to provide reduced-order nonlinear models that can both accurately fit various data sets and computationally simulate new responses is illustrated through studies considering both synthetic data and experimental data. In both cases, the results obtained with the naïve elastic net are shown to match or outperform those from other traditional methods.     

A continuum constitutive model for the mechanical behavior of woven fabrics

We propose a new approach for developing continuum models for the mechanical behavior of woven fabrics in planar deformation. We generate a physically motivated continuum model that can both simulate existing fabrics and predict the behavior of novel fabrics based on the properties of the yarns and the weave. The approach relies on the selection of a geometric model for the fabric weave, coupled with constitutive models for the yarn behaviors. The fabric structural configuration is related to the macroscopic deformation through an energy minimization method, and is used to calculate the internal forces carried by the yarn families. The macroscopic stresses are determined from the internal forces using equilibrium arguments. Using this approach, we develop a model for plain weave ballistic fabrics, such as Kevlar®, based on a pin-joined beam geometry. We implement this model into the finite element code ABAQUS and simulate fabrics under different modes of deformation. We present comparisons between model predictions and experimental findings for quasi-static modes of in-plane loading.     

On one dynamic problem for structurally inhomogeneous beams

A method is developed for studying the dynamic deformation of structurally inhomogeneous beams consisting of homogeneous isotropic layers with different mechanical characteristics. The method is based on the virtual-displacement principle. The equation of motion is derived in vector and scalar forms for arbitrary loads, boundary conditions, and cross-sections with one and two axes of symmetry. The efficiency of the method is demonstrated by solving, as an example, the dynamic deformation problem for a hinged layered beam with a rectangular cross-section under harmonic loading. Mechanical effects are revealed, which describe the influence of the beam structure and the mechanical properties of beam components on the dynamic compliance in comparison with the relevant homogeneous beam with the same geometry __________ Translated from Prikladnaya Mekhanika, Vol. 43, No. 11, pp. 90–98, November 2007.     

Shear deformation effect in non-linear analysis of composite beams of variable cross section

In this paper the non-linear analysis of a composite Timoshenko beam with arbitrary variable cross section undergoing moderate large deflections under general boundary conditions is presented employing the analog equation method (AEM), a BEM-based method. The composite beam consists of materials in contact, each of which can surround a finite number of inclusions. The materials have different elasticity and shear moduli with same Poisson's ratio and are firmly bonded together. The beam is subjected in an arbitrarily concentrated or distributed variable axial loading, while the shear loading is applied at the shear center of the cross section, avoiding in this way the induction of a twisting moment. To account for shear deformations, the concept of shear deformation coefficients is used. Five boundary value problems are formulated with respect to the transverse displacements, the axial displacement and to two stress functions and solved using the AEM. Application of the boundary element technique yields a system of non-linear equations from which the transverse and axial displacements are computed by an iterative process. The evaluation of the shear deformation coefficients is accomplished from the aforementioned stress functions using only boundary integration. Numerical examples are worked out to illustrate the efficiency, the accuracy, the range of applications of the developed method and the influence of the shear deformation effect.