浮筏系统中的碰撞动力学离散模型(1):确定性模型

本文研究浮筏中刚性连结的零部件冲击受损后在有限空间的碰撞振动,应用碰撞动力学理论建立其确定性离散动力学模型,导出确定性系统的碰撞庞加莱映射.实例分析指出,零部件的质量和冲击后外激励的频率不影响零部件与设备之间的碰撞,碰撞主要发生在零部件与筏体之间,而且随着零部件质量的增加碰撞次数发生有限次的增加.冲击后外激励的频率对零部件与筏体之间的碰撞次数有一定的影响.从碰撞庞加莱图可以看出碰撞系统的振动表现为亚谐振动.     

非高斯随机激励下滞迟系统响应与首次穿越失效

针对由有界噪声、泊松白噪声和高斯白噪声共同构成的非高斯随机激励,通过Monte Carlo数值模拟方法研究了此激励作用下双线性滞迟系统和Bouc-Wen滞迟系统这两类经典滞迟系统的稳态响应与首次穿越失效时间。一方面,分析了有界噪声和泊松白噪声这两种分别具有连续样本函数和非连续样本函数的非高斯随机激励,在不同激励参数条件下对双线性滞迟系统和Bouc-Wen滞迟系统的稳态响应概率密度、首次穿越失效时间概率密度及其均值的不同影响;另一方面,揭示了在这类非高斯随机激励荷载作用下,双线性滞迟系统的首次穿越失效时间概率密度将出现与Bouc-Wen滞迟系统的单峰首次穿越失效时间概率密度截然不同的双峰形式。     

非高斯随机激励下滞迟系统响应与首次穿越失效

针对由有界噪声、泊松白噪声和高斯白噪声共同构成的非高斯随机激励,通过Monte Carlo数值模拟方法研究了此激励作用下双线性滞迟系统和Bouc-Wen滞迟系统这两类经典滞迟系统的稳态响应与首次穿越失效时间。一方面,分析了有界噪声和泊松白噪声这两种分别具有连续样本函数和非连续样本函数的非高斯随机激励,在不同激励参数条件下对双线性滞迟系统和Bouc-Wen滞迟系统的稳态响应概率密度、首次穿越失效时间概率密度及其均值的不同影响;另一方面,揭示了在这类非高斯随机激励荷载作用下,双线性滞迟系统的首次穿越失效时间概率密度将出现与Bouc-Wen滞迟系统的单峰首次穿越失效时间概率密度截然不同的双峰形式。     

DYNAMICAL CHARACTERISTICS ANALYSIS FOR UNILATERAL CONSTRAINED SIMPLY SUPPORTED BEAM WITH HARMONIC WAVE EXCITATION1)

为研究间隙碰撞对系统动力学响应的影响,以理想简支梁的振型函数为Rayleigh--Ritz基函数建立了单侧约束简支梁系统的非线性离散动力学方程组,应用数值方法研究了系统在基础谐波激励下的动力学响应特征及其对共振频率线性等效方法适用性的影响。研究表明：非线性动力学系统间隙产生的局部碰撞,使得系统振动能量在系统各阶模态之间转移,使得线性等效方法失效;即使进行非线性分析,也需要考虑系统固有频率远大于激励频段上限的模态。     

基础谐波激励单侧约束简支梁系统的动力学特征分析

为研究间隙碰撞对系统动力学响应的影响,以理想简支梁的振型函数为Rayleigh–Ritz基函数建立了单侧约束简支梁系统的非线性离散动力学方程组,应用数值方法研究了系统在基础谐波激励下的动力学响应特征及其对共振频率线性等效方法适用性的影响。研究表明:非线性动力学系统间隙产生的局部碰撞,使得系统振动能量在系统各阶模态之间转移,使得线性等效方法失效;即使进行非线性分析,也需要考虑系统固有频率远大于激励频段上限的模态。     

冲击载荷下蛋白质水溶液吸附成膜的润滑试验研究

利用球盘接触形式的光干涉膜厚测量试验机,研究了周期冲击载荷作用下蛋白质水溶液在材料表面的吸附膜生长特性.对比了溶液蛋白质组分、亲疏水对偶表面和冲击条件对球盘接触区蛋白质吸附膜厚的影响.结果显示动态冲击下蛋白质的吸附膜厚要远高于静态吸附,吸附膜不断在接触区生长,之后趋于稳定.金属钢球比陶瓷球、疏水盘表面比亲水盘面,均更有利于蛋白质吸附膜的生长.同时发现吸附膜厚与单一蛋白质的浓度关系不大,却和蛋白质种类关系很大,球蛋白比白蛋白具有更好的吸附成膜性能.     

刨削参数对刨刀载荷特性影响研究及其强度分析

为研究刨刀刨削煤岩过程中刨削参数对刨刀载荷的影响,建立了基于碰撞接触模型的刨刀刨削煤岩动力学模型。采用软件模拟、现场试验的手段,对不同工况条件下的刨刀载荷进行了分析,得到刨削速度、刨削深度及煤岩硬度等刨削参数对刀头载荷的影响规律;并基于仿真结果,对刨刀强度进行了分析和预测。研究结果表明:随着刨削深度的增加,刨刀载荷特征值均有较大幅度的增加,刨削速度、煤岩硬度等刨削参数与刨刀载荷特征值也呈正相关关系,但变化幅度较小。根据仿真结果选取一组最恶劣工况条件进行刨刀强度分析,发现除焊缝局部在接触载荷作用下产生强度失效外,其余大部分区域仍可以满足材料强度使用需求,并不会产生失效;在极限测试载荷条件下刨刀焊缝部位也可能发生强度失效,且有较小概率引发刀头、刀体部分发生断裂失效。研究结果可为刨煤机关键零部件设计优化提供一定的理论基础。     

浮筏系统中的碰撞动力学离散模型(2):随机模型

本文建立了浮筏系统在部分冲击受损后形成的碰撞振动系统的离散动力学随机模型.在此随机模型中,浮筏系统的运动满足高斯分布,零部件的运动方程中的非线性是由于空隙单元产生,可以采用非高斯截断技术导出随机系统的平均碰撞庞加莱映射.实例分析指出,随机模型中由于增加了噪声的摄动,表现出的动力学特性比确定性模型更丰富,噪声的影响有时会增加零部件与设备之间的碰撞次数,也会抑制零部件与筏体之间的碰撞次数,噪声将增加系统动力学特性的复杂性.     

多种冲击载荷条件下的人体肋骨骨折有限元分析

旨在研究不同碰撞载荷条件下基于不同失效模型的人体肋骨骨折机理. 为此采用已验证的人体有限元胸部模型来分析人体肋骨骨折现象. 该模型基于人体解剖学结构,包含了人体胸椎、腰椎、肋骨、胸骨、肋间软骨、胸腹部器官和其他的软组织,定义的生物材料参数都基于已有的文献记载. 基于人体在损伤生物力学领域内一些较为典型的肋骨骨折失效模型,根据已发表文献中的人体标本实验载荷条件模拟了人体肋骨结构在不同冲击载荷下的骨折现象,并与这些实验结果进行了对比分析. 所引用的实验结果包括单根肋骨强度结构实验和人体胸部正面碰撞块冲击实验. 从文中有限元仿真分析的结果来看,针对不同的载荷条件,不同肋骨骨折失效模型的适用性各不相同. 该人体胸部有限元模型可用于车辆交通事故中冲击载荷条件下的人体肋骨损伤生物力学研究.     

基于区域分解的结构动力学系统首次穿越失效

提出了高斯白噪声激励的线性及非线性结构动力学系统的首次穿越失效概率的估计方法. 对于线性结构动力学系统,失效区域被分解为互斥的基本失效域之和,每个基本失效域可用其设计点完全描述,并以正态分布代替卡方分布估计失效概率中的参数. 对于非线性结构动力学系统,基于Rice穿越理论,将非线性方程转化为与之具有相同平均上穿率的线性化方程,然后利用文中方法对等效线性化方程估计首穿失效概率. 最后给出了线性及非线性结构动力学系统的数值例子,并将所提方法与蒙特卡罗法及重要样本法相比较,模拟结果显示了方法的正确性与有效性.     

Computing the aeroelastic disk vibrations in a hard disk drive

The turbulent flow of air caused by the spinning of a single disk inside a typical hard disk drive casing is calculated using large eddy simulation (LES). The pressure acting on the disk is recorded as a function of time and is used to compute the vibrations of the spinning disk using a self-developed hybrid-spectral finite-difference code. This unidirectional fluid–structure interaction problem is computed for two commonly occurring cases: a disk actuated on one side only (Case 1) and a disk actuated on both sides (Case 2). The pressure loading on the disk is characterized in terms of its mean, root-mean-square (r.m.s.) and its spectral content. The mean pressure acting on the disk is asymmetrical in the case where the disk is actuated on one side only, leading to a mean deformation of the disk to one side. The r.m.s. vibrations of Case 2 are higher than those for Case 1 and their spectral distributions are almost identical. Large pressure fluctuations of the flow are found in the wake of the actuator arm and near the region where the shroud expands to accomodate the actuator. The spectral content of the excitation force due to the pressure is mainly in the low kHz frequency range, while higher frequencies are seen at the disk edge. This typically results in the excitation of the first 3–4 modes of the disk; however, (asymmetric) Case 1 displays the excitation of higher modes compared with (symmetric) Case 2.     

Modeling of hard disk drives for shock and vibration analysis – consideration of nonlinearities and discontinuities

This paper reviews recent advances (mostly after year 2000) in shock and vibration analysis of hard disk drives (HDD) considering the presence of nonlinearities and discontinuities. Components and dynamic phenomena in HDD where effects of mechanical nonlinearity and discontinuities are significant are discussed, e.g., head actuator suspension, dimple and slider, head–disk interface, fluid dynamic bearing, spinning disks, and load/unload dynamics. Ways to model these nonlinearities and discontinuities are reviewed in detail. Our research on modeling an entire HDD in operating mode subject to shock and vibration using a flexible multibody dynamics formulation is also presented. The numerical simulation of the shock response of a 1-in. form factor HDD is presented. A half-sinusoidal acceleration shock is applied at the base of the HDD. Response of the flying height for different shock amplitudes and duration times is simulated.     

Simultaneous measurements of disk vibration and pressure fluctuation in turbulent flow developing in a model hard disk drive

The complex flow features inside hard disk drive models are investigated in an axisymmetric and a semi-open shroud configurations. For the axisymmetric case, we have employed both experimental and computational approaches. The experiment focuses on both flow dynamics and the disk vibration, where measurements of the fluctuating pressure and velocity are undertaken at some representative points. The correlation between the disk vibration and the fluctuating pressure in the turbulent flow between disks is evident from the spectral analysis. The experimentally observed fluctuating pressure and velocity are partly due to the disk vibration and its contribution could be estimated by comparing the experiment with the results of a large eddy simulation. For the semi-open shroud case, although the characteristic peaks attributable to the large-scale vortical structure are still observed in the power spectra, the pressure fluctuation and the disk vibration are suppressed when the arm is inserted.     

Primary wave transmission in systems of elastic rods with granular interfaces

We study analytically and numerically primary pulse transmission in one dimensional systems of identical linearly elastic non-dispersive rods separated by identical homogeneous granular layers composed of n beads. The beads interact elastically through a strongly (essentially) nonlinear Hertzian contact law. The main challenge in studying pulse transmission in such strongly nonlinear media is to analyze the ‘basic problem’, namely, the dynamical response of a single intermediate granular layer, confined from both ends by barely touching linear elastic rods subject to impulsive excitation of the left rod. The analysis of the basic problem is carried out under two basic assumptions; namely, of sufficiently small duration of the shock excitation applied to the first layer of the system, and of sufficiently small mass of each bead in the granular interface compared to the mass of each rod. In fact, the smallness of the mass of the bead defines the small parameter in the asymptotic analysis of this problem. Both assumptions are reasonable from the point of view of practical applications. In the analysis we focus only in primary pulse propagation, by neglecting secondary pulse reflections caused by wave scattering at each granular interface and considering only the transmission of the main (primary) pulse across the interface to the neighboring elastic rod. Two types of shock excitations are considered. The first corresponds to fixed time duration (but still much smaller compared to the characteristic time of pulse propagation through the length of each rod), whereas the second type corresponds to a pulse duration that depends on the small parameter of the problem. The influence of the number of beads of the granular interface on the primary wave transmission is studied, and it is shown that at granular interfaces with a relatively low number of beads fast time scale oscillations are excited with increasing amplitudes with increasing number of beads. For a larger number of beads, primary pulse transmission is by means of solitary wave trains resulting from the dispersion of the original shock pulse; in that case fast oscillations result due to interference phenomena caused by the scattering of the main pulse at the boundary of the interface. Considering a periodic system of rods we demonstrate significant reduction of the primary pulse when transmitted through a sequence of granular interfaces. This result highlights the efficacy of applying granular interfaces for passive shock mitigation in layered elastic media.     

Pulse shaping techniques for testing brittle materials with a split hopkinson pressure bar

We present pulse shaping techniques to obtain compressive stress-strain data for brittle materials with the split Hopkinson pressure bar apparatus. The conventional split Hopkinson pressure bar apparatus is modified by shaping the incident pulse such that the samples are in dynamic stress equilibrium and have nearly constant strain rate over most of the test duration. A thin disk of annealed or hard C11000 copper is placed on the impact surface of the incident bar in order to shape the incident pulse. After impact by the striker bar, the copper disk deforms plastically and spreads the pulse in the incident bar. We present an analytical model and data that show a wide variety of incident strain pulses can be produced by varying the geometry of the copper disks and the length and striking velocity of the striker bar. Model predictions are in good agreement with measurements. In addition, we present data for a machineable glass ceramic material, Macor, that shows pulse shaping is required to obtain dynamic stress equilibrium and a nearly constant strain rate over most of the test duration.     

Highly spatially resolving laser Doppler velocity measurements of the tip clearance flow inside a hard disk drive model

The flow in the tip clearance of a hard disk drive model has been investigated with laser Doppler techniques. The flow was driven by co-rotating disks inside a cylindrical enclosure in order to simulate a hard disk drive used for data storage devices. The main focus of the investigation was on the understanding of complex flow behavior in the narrow gap region between the disk tip and the outer shroud wall, which is supposed to be one of the causes of flow induced vibration of the disks. Experiments in the past have never been able to examine this region because of the lack of the spatial resolution of sensors in the highly three-dimensional flow in the region. In the present investigation, the flow velocity in the tip clearance region was measured with optical measurement techniques for the first time. The flow behaviors are investigated for four different conditions with two different gap widths and two different shapes of the shroud walls with and without ribs. The velocity measurements were taken both with conventional laser Doppler velocimetry and using a laser Doppler velocity profile sensor with a spatial resolution in the micrometer range. The circumferential velocity component was measured along the axial and radial directions. The steep gradients of the circumferential mean velocity in both directions were successfully captured with a high spatial resolution, which was achieved by the velocity profile sensor. From the supplementary investigations, the existence of vortex structures in the tip clearance region was confirmed with a dependence on the shroud gap width and the shroud shape. The interactions of the two boundary layers seem to be the source of the complex three-dimensional behaviors of the flow in this region.     

超高车辆与箱梁立交桥碰撞简化计算模型评估分析

为了考察受超高车辆撞击装配式钢筋混凝土箱梁跨线桥的冲击动力和破坏行为，以一起近来发生的实际工程事故为案例进行精化有限元数值分析，并提出了双质量-并联弹簧（double mass-parallel spring, DM-PS）简化车辆模型，以有效地模拟超高车辆与桥梁的非对心碰撞行为。所建议DM-PS简化模型的有效性通过与两种广泛使用的车辆模型包括全尺（full scale, FS）模型和简单刚体（simple rigid, SR）模型的比较而得到充分地评估。计算结果表明：采用FS模型可得到与事故现场照片基本一致的跨线桥撞击区域破坏特征；SR模型高估结构的局部破坏，弱化结构的整体变形；DM-PS模型对于预测结构破坏具有较高的准确性。因此，所提出的DM-PS模型为超高车辆撞击桥梁结构防护设计提供了一个简单有效的分析手段。在此基础上，利用DM-PS模型进行了详细的结构行为参数分析，深入考察了车辆撞击速度、撞击质量、撞击位置以及结构形式等效应。所得到的结论为：相比撞击质量，结构的冲击动力行为对于撞击速度有更高的敏感性；跨中受撞和边跨受撞的变形和破坏模式有较大差异，边跨受撞对于单侧支座损伤更严重；...     

Dynamics of stress wave propagation in a chain of photoelastic discs impacted by a planar shock wave; Part I, experimental investigation

The propagation of stress waves through a chain of discs has been studied experimentally. Optically transparent 20-mm diameter discs, made of epoxy, were loaded dynamically by head-on collision with an incident planar shock wave. The loading was done in a vertical shock tube. The head-on collision between the punch-plate, placed on top of the chain of discs, and the incident shock wave resulted in a head-on reflected shock wave inducing behind it a fairly uniform step-wise pressure pulse having duration of about 6 ms. The recorded fringe patterns of the stress field, in the discs-chain, show that the input pressure pulse was broken into several oscillating cycles. The back and forth bouncing of stress waves gave rise to two different modes of the contact stress oscillations, which continued until the overall stress reaches equilibrium with the input conditions. The registered propagation velocity of the stress wave was significantly lower than the appropriate speed of sound in the material from which the discs were made.      

Dynamic buckling of a shallow arch under shock loading considering the effects of the arch shape

In this paper, the influence of the initial curvature of thin shallow arches on the dynamic pulse buckling load is examined. Using numerical means and a multi-dof semi-analytical model, both quasi-static and non-linear transient dynamical analyzes are performed. The influence of various parameters, such as pulse duration, damping and, especially, the arch shape is illustrated. Moreover, the results are numerically validated through a comparison with results obtained using finite element modeling. The main results are firstly that the critical shock level can be significantly increased by optimizing the arch shape and secondly, that geometric imperfections have only a mild influence on these results. Furthermore, by comparing the sensitivities of the static and dynamic buckling loads with respect to the arch shape, non-trivial quantitative correspondences are found.     

Simulating radiative shocks with the CRASH laser package

We present the latest improvements in the Center for Radiative Shock Hydrodynamics (CRASH) code, a parallel block-adaptive-mesh Eulerian code for simulating high-energy-density plasmas. The implementation can solve for radiation models with either a gray or a multigroup method in the flux-limited-diffusion approximation. The electrons and ions are allowed to be out of temperature equilibrium and flux-limited electron thermal heat conduction is included. We have recently implemented a CRASH laser package with 3-D ray tracing, resulting in improved energy deposition evaluation. New, more accurate opacity models are available which significantly improve radiation transport in materials like xenon. In addition, the HYPRE preconditioner has been added to improve the radiation implicit solver. With this updated version of the CRASH code we study radiative shock tube problems. In our set-up, a 1 ns, 3.8 kJ laser pulse irradiates a 20 micron beryllium disk, driving a shock into a xenon-filled plastic tube. The electrons emit radiation in the shocked xenon. This radiation preheats the unshocked xenon. Photons traveling ahead of the shock will also interact with the plastic tube, heat it, and in turn this can drive another shock off the wall into the xenon.