基于代理模型的箭体连接结构极值响应分析与优化

螺栓法兰连接结构在航空航天等工程领域中广泛应用，其力学性能在不同工况和装配情况下十分复杂。由于拉压刚度差异，含连接结构的箭体动力学响应呈现明显的非线性特征。因此，考虑不同连接参数及工况下的连接非线性动力学响应，对结构优化设计有着重要意义。本文针对以双线性弹簧表征螺栓法兰连接非线性的箭体等效动力学模型，基于径向基函数(RBF)神经网络和响应面法分别建立其连接面处的极值响应代理模型，对比发现RBF神经网络模型在较高精度上可以实现对动响应极值的预测及分析；同时分析了不同载荷参数及刚度变化对连接结构动响应极值的影响；最后，利用RBF神经网络代理模型，开展了连接面加速度极值响应与螺栓弹簧力最小化为目标的连接结构参数优化。     

螺栓连接大变形梁振动特性的实验与数值研究

螺栓连接的柔性梁既有非线性又有几何非线性,本文针对其动力学建模和振动特性开展了实验与数值研究。首先,搭建了含螺栓连接的柔性大变形梁的实验台架,并开展敲击和正弦激振的实验测试。实验结果表明:螺栓连接的柔性梁较(无螺栓连接的)连续梁的模态频率降低,阻尼增加,呈现出随着激励能量增大,模态频率降低的非线性模态特征。其次,为了建立大变形柔性梁的有效计算模型,采用了动坐标迭代法,并采用改进Iwan模型和库仑摩擦模型刻画连接部件的力学特性。最后,开展了数值计算,通过与实验测试结果对比表明:数值计算结果与实验测试结果误差在7%以内,所建立的柔性连接梁的计算模型能够有效准确地模拟含连接部件柔性梁的非线性模态和动力学响应特征。     

基于六参数非均匀密度函数的伊万模型研究

伊万模型(Iwan model) 常用于描述连接的能量耗散和阻尼特性,来研究连接结构的非线性静动力学行为.本文提出了一个含截断幂律分布和双脉冲的非均匀密度函数的六参数伊万模型, 推导出六参数伊万模型的骨线方程,微、宏观滑移情况下的卸载方程和能量耗散的解析表达式,并讨论了模型参数对能量耗散的影响. 与已有的伊万模型相比,六参数伊万模型能够更好地描述连接结构的实验现象.      

柔性梁刚柔耦合碰撞动力学及变形传播研究

运用柔性多体系统刚柔耦合动力学理论,研究了作大范围回转运动柔性梁的碰撞动力学问题.考虑柔性梁的横向变形,以及横向变形引起的纵向缩短项即非线性耦合变形项.采用基于Hertz接触理论及非线性阻尼理论的非线性弹簧阻尼模型来求解碰撞过程中产生的碰撞力,运用第二类拉格朗日方程建立了系统的刚柔耦合碰撞动力学方程.编制仿真软件进行动力学仿真计算,得到了碰撞力和系统动力学响应,对比分析了不同动力学模型对系统动力学响应的影响.同时研究了碰撞导致的柔性梁横向变形传播的波动特性.     

连接结构接触界面非线性力学建模研究

连接界面上存在的跨尺度、多物理场和非线性行为是引起结构复杂非线性动力学的主要原因.由于连接界面的力学行为的复杂性,以及难以对连接界面进行直接试验观测,连接界面的力学建模一直是非常具有挑战性的科学问题.本文首先从分析结合面的跨尺度物理机理入手,将名义的光滑平面视作凹凸不平的粗糙面,考虑单个微凸体的黏滑摩擦行为,建立接触载荷与变形的非线性关系,然后采用GW(Greenwood and Williamson)模型数理统计方法建立整个粗糙界面的跨尺度力学模型,并与公开文献中试验结果进行对比.考虑连接界面具有典型非线性特征,提出一种改进的Iwan唯象模型,利用精细有限元方法获得非线性特征结果,采用系统辨识理论建立连接结构的降阶力学模型,并利用有限元结果进行模型验证.结果表明,本文提出的粗糙界面跨尺度模型在法向载荷较小时与试验结果吻合较好,改进的Iwan模型能够较好描述连接界面的非线性特征,并与有限元结果吻合较好.     

弹性波在螺栓搭接结构中传播行为的数值模拟

为了建立有效的基于弹性波的螺栓连接损伤检测方法,研究弹性波在连接结构中的传播行为至关重要。本文针对典型单螺栓搭接薄梁,采用非线性多物理场耦合有限元仿真方法,研究弹性波在连接结构中的传播行为。采用ANSYS中压电耦合单元建模MFC压电传感器,非线性接触单元建模螺栓连接结构,将接触面之间考虑为考虑摩擦接触的非线性关系。依次对螺栓施加预紧载荷,对结构施加模拟电压激励,进行非线性瞬态波传播分析。同时,本文对仿真结果进行了实验验证。结果表明,本文建立的有限元模型和分析方法能够有效地模拟弹性波在连接结构中的传播过程,数值预测的响应电压和实验结果吻合较好。针对非线性模型计算耗费大的问题,本文对非线性模型进行了线性化改进。具体为采用准静态接触有限元分析确定对接面有效接触面积的大小,然后将接触面积上对应节点绑定,进行线性瞬态波传播分析。结果表明:本文的线性方法能够有效模拟波传播过程,并能有效降低计算耗费;弹性波在连接梁中传播时,经过连接界面和螺栓位置时,传播路径、波形、幅值都发生了较大的变化。     

连接结构接触界面非线性力学建模研究

连接界面上存在的跨尺度、多物理场和非线性行为是引起结构复杂非线性动力学的主要原因。由于连接界面力学行为的复杂性,以及对连接界面进行直接试验观测的困难,连接界面的力学建模一直是非常具有挑战性的科学问题。本文首先从分析结合面的跨尺度物理机理入手,将名义的光滑平面视作凹凸不平的粗糙面,考虑单个微凸体的黏滑摩擦行为,建立接触载荷与变形的非线性关系,然后采用GW(Greenwood和Williamson, GW)模型数理统计方法建立整个粗糙界面的跨尺度力学模型,并与公开文献中试验结果进行对比。然后考虑连接界面典型非线性特征,提出一种改进的Iwan唯象模型,利用精细有限元方法获得非线性特征结果,采用系统辨识理论建立连接结构的降阶力学模型,并利用有限元结果进行模型验证。结果表明,本文提出的粗糙界面跨尺度模型在法向载荷较小时与试验结果吻合较好,改进的Iwan模型能够较好描述连接界面非线性特征,并与有限元结果吻合较好。      

伴随变阻尼作用的干摩擦下的车辆系统非线性动力学分析

对分段线性阻尼和干摩擦共同作用下的车辆悬挂系统进行了非线性动力学分析研究，阐述了判定系统周期运动稳定性的理论方法；利用数值模拟方法分析了具有不同阻尼参数组合的系统对简谐激励的振动响应，并分析了由干摩擦引起的粘-滑振动行为．结果表明：提高摩擦力对抑制响应有利，但车辆系统在低速下运行时会出现复杂的粘-滑振动，轮轨之间产生较大的瞬时刚性冲击；而通过增加轮对与侧架的弹性悬挂可以有效减弱这种瞬时刚性冲击．     

结构阻尼模型及系统时域动响应

振动界对经典的非频变结构阻尼模型一直有种种争议,本文检查了它的前提并指出将该模型推广到系统时域动响应分析时导致的悖论.为了提供结构阻尼系统的时域动响应实用分析方法,文中分别提出了基于响应优势谱成份的粘性阻尼近似模型和基于粘弹性理论的三参数频变结构阻尼模型;它们既可逼近结构阻尼缓频变的特性,又可保证系统的时域动响应具有因果性.     

基于子结构解耦的连接特性识别新方法

连接部件动态特性的准确辨识对预测装配式机械结构的动力学行为有重要意义. 针对传统基于子结构解耦的连接结构动力学特性识别方法难以直接利用可测量数据进行辨识及辨识结果受噪声影响显著等问题, 本文提出了一种新方法. 首先, 提取子结构解耦基本方程在测试自由度上的分量, 并经矩阵变换得到显含连接动刚度矩阵的形式, 而后由真实连接动刚度矩阵分解为已知的初始矩阵与待求的增量矩阵, 推导了具有收敛性质的增量型方程以增强界面自由度较多时辨识的数值稳定性, 并采用多项式拟合动刚度将其转化为了拟合系数的频域估计方程, 按给定准则选取合适的频率点联立方程后, 得到了只需装配体测试自由度上的频响函数来辨识连接特性的迭代公式. 最后, 以若干算例说明了算法的具体流程. 对10自由度弹簧?质量块系统进行了数值仿真, 验证了所提方法的正确性及抗噪性; 对包含一处胶接连接的T形梁结构和包含两处螺栓连接的L形梁结构进行了试验, 所辨识连接结构与残余结构重组的装配体有限元模型计算的频响函数与测量值在较宽频带内吻合较好, 表明了该方法能有效识别实际装配体结构中的连接特性.      

两种典型螺栓连接结构的刚度特性比较研究

针对法兰对接和径向套接两种典型舱段螺栓连接形式,基于有限元静刚度计算、动力学简化建模、冲击响应特性分析及结构冲击试验,系统研究了两种连接形式的轴向刚度特性及其对动力学冲击响应的影响.有限元静刚度分析揭示了法兰对接与径向套接的轴向拉压刚度分别是非对称的和对称的,而法兰对接的平均刚度更大.之后,为两种连接形式建立了统一的动力学模型,证明非对称的轴向拉压刚度导致结构在受到横向载荷作用时会产生附加的耦合轴向振动,并且利用高精度幂函数拟合刚度跳变,得到耦合轴向振动频率是弯曲振动频率的二倍的结论.最后,通过冲击动力学试验证明了法兰对接存在二倍频的耦合轴向振动,而径向套接则不存在该耦合振动.径向套接虽然一阶频率较低,但阻尼效果更好.     

几种典型钢框架节点连接性能的有限元仿真分析

采用有限元程序ABAQUS进行仿真分析,探讨了钢框架节点的非线性连接性能。针对几种典型节点,包括焊接连接、外伸式端板螺栓连接、平齐式端板螺栓连接、双腹板顶底角钢连接等,分别建立了相应的精化有限元分析模型,实现了结构非线性行为的真实再现和连接性能的准确评估。数值结果还给出了钢框架节点的弯矩-相对转角、弯矩-剪切转角及弯矩-节点转角三种关系曲线,并针对弯矩-节点转角曲线进行了回归分析,从而得到了四种钢框架节点的弯矩-节点转角关系式。建立的有限元分析模型可真实地模拟钢框架节点的非线性行为,分析得到了半刚性节点弯矩-转角关系的一般规律,可为钢框架梁柱连接节点的计算理论和设计方法提供重要参考和依据。     

Three-dimensional investigation of thick single-lap bolted joints

Stresses in single-lap bolted joints of thick plates are complex and difficult to analyze. Previous studies involving stresses through the thickness of bolted joints have been limited to finite element method (FEM) simulations and have been implemented only for the joining of relatively thin plates. In this paper we report on several experimental and numerical analyses that were conducted to evaluate the stress distribution inside thick bolted plates along the bearing plane normal to the plate surface. Experimental analysis was conducted via embedded-polariscope photoelasticity and embedded resistance strain gages. The FEM analysis was performed with the ABAQUS commercial code using material properties and other data obtained experimentally as input. Experimental and numerical results agreed reasonably well, and are believed to depict the behavior of the joint under load well enough to assist in development of improved joint design.     

Monitoring of L-Shape Bolted Joint Tightness Using Thermal Contact Resistance

In this paper, we describe the possible use of hermal contact resistance measured under ambient conditions as a non-destructive feature to monitor L-shape bolted joints. We demonstrate that thermal contact resistance can be correlated to the contact pressure of L-shape bolted joints and thus can be used to infer joint tightness. Five different torque levels of bolt tightness in an L-shape bolted joint were measured for contact pressure and thermal contact resistance using Fuji Prescale® pressure-sensitive film and a test setup in ambient conditions. Using probabilistic analysis, probability density functions (PDFs) and intervals developed from the PDFs, the contact pressure and thermal contact resistance were developed and correlated. By examining the intervals of the contact pressure and thermal contact resistance at each torque level, it was concluded that thermal contact resistance measured under ambient conditions can be used to describe bolted joint tightness.     

机匣安装边结构研究

本文利用多种实验力学方法研究了航空发动机机匣安装边结构的连接特性和力学行为以及机匣在安装，受载过程中的安装边的变形、应力和螺栓内力、应力等的规律；讨论了安装边结构设计参数的选择范围     

Application of subharmonic resonance for the detection of bolted joint looseness

Bolted joint structures are prone to bolt loosening under environmental and operational vibrations, which may severely affect the structural integrity. This paper presents a bolt looseness recognition method based on the subharmonic resonance analysis. The bolted joint structure was simplified to a two-degree-of-freedom nonlinear model, and a multiple timescale method was used to explain the phenomenon of the subharmonic resonance and conditions for the generation of subharmonics. Numerical simulation predictions for the generation of the subharmonics and conditions for the subharmonics can be found with respect to the excitation frequency and the excitation amplitude. Experiments were performed on a bolt-joint aluminum beam, where the damage was simulated by loosening the bolts. Two surface-bonded piezoelectric transducers were utilized to generate continuous sinusoidal excitation and to receive corresponding sensing signals. The experimental results demonstrated that subharmonic components would appear in the response spectrum when the bolted structure was subjected to the excitation of twice its natural frequency. This subharmonic resonance method was found to be effective on bolt looseness detection.     

A method to simplify the strength design of bolted joints—Case of connecting-rod bolts

This study shows the development of the strength-design system for general bolted joints. Specifically, a simplified method of the strength design for the connectingrod bolts is proposed. The connecting-rod joint is transformed into a single-bolted-joint model by using the ring model. The axial stress on the bolt and separation behavior at the contact plane are discussed. The strength design of the connecting-rod joint can be performed by applying the unified procedure to the single-bolted-joint model. The local deformation of the connecting-rod joint and the stress on the bolt were measured and the analytical results are compared with the experimental results. Both results show fairly good agreement.     

Definition and updating of simplified models of joint stiffness

The objective of this work is to define a simple linear model of joints used in aeronautics and to update this model efficiently.Industrial designers usually resort to semi-empirical linear joint models to represent the behavior of the joints of a large aeronautical structure. Here, we propose to develop a one-dimensional linear joint model which is capable of representing the behavior of every joint of a large structure globally while enabling local nonlinear reanalysis of the most highly loaded joints. Work on nonlinear reanalysis is not considered in this paper.In order to solve the numerical difficulties encountered in some of modeling situations, an updating strategy based on the constitutive relation error is proposed. Since the updating efficiency is significantly affected by the ratios of the stiffnesses of the different parts of the model, the strategy consists in rigidifying some parts of the model in order to control the updating accuracy and the rate of convergence. The numerical results of a standard model and a rigidified model illustrate the updating improvements allowed by the strategy.     

Fatigue design of bolted joints taking into account reliability concepts

A method is presented for fatigue design by comparing the load on the bolt in joints with the fatigue strength of the bolt-nut joints. Two types of bolted joints, cylindrical and T-flange joints, are treated as examples. Reliability concepts are introduced to the design. Fatigue tests of the above-mentioned bolted joints were also carried out to verify the applicability of the method. As a result, it appears that the fatigue behavior of threaded fasteners can be estimated to some extent by comparing the load on the bolt with the fatigue strength. Paper was presented at the 1986 SEM Spring Conference on Experimental Mechanics held in New Orleans, LA on June 8–13.     

Analytical parametric study of bi-linear hysteretic model of dry friction under harmonic,impulse and random excitations

Hysteretic behavior due to some nonlinear sources is a common phenomenon in many dynamical systems. One of the sources of this behavior in mechanical systems is dry friction. Dry friction in bolted or riveted joints of mechanical structures makes their dynamic behavior hysteretic. Bi-linear hysteresis is one of the models that can be used to study these systems which is used in this paper. A SDOF system containing a bi-linear hysteretic element called Jenkins element under harmonic, impulse and random excitations is considered. For all three types of excitations, the effects of system and excitation parameters on the defined equivalent system parameters and the response specifications are studied. Harmonic balance method is employed for harmonic excitation studies, and optimum friction threshold for minimizing response amplitude is obtained versus other system parameters and response amplitude. Energy balance method is used for impulse excitation through which the desired decaying ratio can be achieved by tuning the friction threshold, depending on stiffness ratio. System under random excitation is investigated by equivalent linearization technique in two steps. At the first step, equivalent properties are obtained versus instantaneous amplitude of response. In this step, the paper contains the parametric study of system in which the variations of equivalent parameters are described when physical parameters of system or input intensity vary. Overall variance of system response is determined in the second step, and optimum sliding threshold is obtained to have minimum overall variance of system response.