单向相关下的机械零件动态可靠性分析

对于载荷与强度存在相关性的情形,应用传统的应力-强度干涉模型具有一定的局限性。为评估零件在应力与强度单向相关下的动态可靠度,利用Gamma过程来描述随机载荷作用,基于经典的Miner累积损伤理论,推导随着载荷作用而产生的强度不断降低的强度退化模型。结合应力-强度干涉模型,进而建立零件的动态可靠度模型。最后通过算例验证模型的有效性,结果表明,该模型能够有效用于零件在单向相关下的动态可靠性评估。     

载荷作用次数和强度退化对SSI模型的影响

在考虑循环载荷作用下强度随使用时间不断退化时,研究了应力-强度干涉模型下的可靠性统计推断问题,给出了一个考虑载荷作用次数应力-强度退化干涉失效模型.应用应力-强度干涉可靠性分析模型进行实例分析计算,结果与实验值符合很好,说明所提出新模型是合理有效的,更符合实际情况.     

随机恒幅循环载荷疲劳可靠度异量纲干涉模型

用于与时间无关的失效模式(例如静强度失效)的失效概率或可靠度计算的"应力-强度干涉模型"已很成熟,但要将干涉分析的概念与方法应用于与时间相关的失效模式(例如疲劳、磨损和腐蚀等)还存在许多困难,即使对于存在不确定性的恒幅循环载荷下的疲劳可靠性问题,目前也还没有像静强度应力-强度干涉模型那样简单、直接、有效的方法及数学模型.在概率统计平均的意义上重新解释传统的两个随机变量干涉分析的基本概念及模型,将干涉模型解释为载荷加权平均模型.具体地讲,就是将应力-强度干涉模型解释、拓展为强度超越(载荷)概率的(载荷)统计加权平均模型,或给定应力下的条件失效概率的随机载荷加权平均模型.这样,传统上只能应用于相同量纲随机变量(例如应力与强度)的干涉模型就可以拓展应用于具有不同量纲随机变量(例如应力与寿命)的情形,例如随机载荷下的疲劳失效概率(寿命小于指定值的概率)或可靠度(寿命大于指定值的概率)计算.应用这样的模型,可以很方便地根据几个确定性恒幅循环载荷下的疲劳寿命分布,预测随机恒幅循环载荷作用下的疲劳失效概率或可靠度.     

应力-强度相关性干涉下的随机安全系数与零件可靠性设计

为解决机械零件可靠度计算中的应力-强度相关性干涉问题,根据两者负相关结构,运用Copula函数的相关性理论,基于零件可靠性计算的相关性干涉模型,建立了应力-强度相关性干涉下的随机安全系数—可靠度计算模型,并给出机械零件的平均安全系数可靠度计算方法.以受拉零件为代表,借助于相关性干涉的联结方程,给出了应力-强度相关性干涉下静强度可靠性设计的通用型求解方法和步骤,通过算例,说明了该模型的有效性、准确性.     

考虑载荷作用次数的零部件可靠性模型

研究随机载荷多次作用时零件的失效过程,建立载荷多次作用下的零件可靠性模型.从载荷作用的统计学意义出发,根据顺序统计量的性质,建立载荷多次作用时等效载荷的累积分布函数和概率密度函数.运用应力-强度干涉模型建立零件在单一失效模式下载荷多次作用时的可靠性模型.最后,运用Monte Carlo方法对可靠性模型进行仿真实验,证明模型的正确性.     

考虑强度退化的零件及系统可靠性分析

指出传统载荷-强度干涉模型适用于载荷作用一次或者静态的零件可靠度计算.零件的强度是随机载荷大小和作用次数的函数,所以假设载荷和强度相互独立的干涉模型无法用于计算考虑强度退化时零件及系统的可靠度.通过随机载荷的统计特性建立强度与随机载荷的关系.当载荷历程为泊松过程时,分析零件的可靠性和失效率与时间的关系,并建立考虑共因失效时串联系统和并联系统的可靠性模型.研究表明,零件初始强度的分散性对零件的可靠度和失效率有很大影响,载荷的相关性使并联系统的可靠度降低,使串联系统的可靠度升高.     

极限工况下风力机主轴强度的可靠性分析

主轴是风力机传动链系统中的关键部件之一,其可靠性直接影响整个机组能否正常工作.基于有限元分析法建立了主轴有限元模型,依据应力强度干涉理论和可靠性分析理论建立了主轴强度可靠性模型.根据主轴各极限工况载荷计算了主轴静强度,分析出其应力-强度分布并计算其可靠度值.将可靠度计算结果与主轴设计可靠度目标分配值进行对比,表明该主轴强度的可靠性可以满足预定要求.     

具有不确定性的恒幅循环载荷疲劳可靠度异量纲干涉分析方法

在概率统计平均的意义上重新解释传统的两个随机变量干涉分析的基本概念及模型,将干涉模型解释为载荷加权平均模型.具体地讲,是将应力一强度干涉模型解释、拓展为强度超越载倚概率的(载荷)统计加权平均模型,或作为应力水平函数的条件疲劳失效概率的载荷统计加权平均模型.这样,传统上只能应用于相同量纲随机变量(例如应力与强度,或载荷循环数与疲劳寿命)的干涉模型可以拓展应用于具有不同量纲随机变量(例如应力与寿命)的情形,即具有不确定性的循环载荷下的疲劳失效慨率(寿命小于指定值的慨率)或可靠度(寿命大于指定值的概率)计算.应用这样的模型,可以很方便地根据几个确定性恒幅循环载荷下的疲劳寿命分布预测具有不确定性的恒幅循环载荷作用下的疲劳失效概率或可靠度.文中还证明,先由寿命分布推导给定寿命下的疲劳强度分布,再根据应力-强度干涉模型计算可靠度的传统公式本质上与文中提出的统计平均模型是一致的.     

多种失效模式下的机械零件动态可靠性模型

考虑共因失效和随机载荷作用次数对可靠性的影响,建立了零件在多种失效模式下的动态可靠性模型。指出共因失效同样存在于零部件的失效中,运用应力-强度干涉模型和顺序统计量理论建立了多种失效模式下载荷多次作用时的零件可靠性模型以及零件动态可靠性模型。以齿轮的可靠度计算为例,研究了零件可靠度和失效率随时间的变化规律。研究表明,当零件强度不退化或退化不明显时,零件的可靠度随时间逐渐降低,失效率曲线具有浴盆曲线前两个阶段的特征。     

考虑失效相关的k/n系统动态可靠性模型

运用载荷-强度干涉模型、顺序统计量理论和概率微分方程建立随机载荷作用下的失效相关k/n系统动态可靠性模型,研究系统可靠度和失效率随时间的变化规律.在不作失效独立假设的前提下,运用系统级载荷-强度干涉模型建立失效相关k/n系统可靠性模型,以及k/n系统强度的累积分布函数和概率密度函数.从随机载荷作用的统计学意义出发,运用顺序统计量建立载荷多次作用下的k/n系统可靠性模型.运用泊松随机过程描述载荷的作用过程,分别建立强度不退化和强度退化时的k/n系统动态可靠性模型,并研究k/n系统可靠度和失效率随时间的变化规律.研究表明,k/n系统的可靠度随时间逐渐降低;k的取值越大,k/n系统的可靠性越低,失效率越高.强度不退化时,k/n系统的失效率随时间逐渐减小且早期失效率较高;强度退化时,k/n系统的失效率具有"浴盆"曲线的特征.     

Discrete Stress-strength Interference Model of Reliability Analysis under Multi-operating Conditions

The conventional stress-strength interference(SSI) model is a basic model for reliability analysis of mechanical components. In this model, the component reliability is defined as the probability of the strength being larger than the stress, where the component stress is generally represented by a single random variable(RV). But for a component under multi-operating conditions, its reliability can not be calculated directly by using the SSI model. The problem arises from that the stress on a component under multi-operating conditions can not be described by a single RV properly. Current research concerning the SSI model mainly focuses on the calculation of the static or dynamic reliability of the component under single operation condition. To evaluate the component reliability under multi-operating conditions, this paper uses multiple discrete RVs based on the actual stress range of the component firstly. These discrete RVs have identical possible values and different corresponding probability value, which are used to represent the multi-operating conditions of the component. Then the component reliability under each operating condition is calculated, respectively, by employing the discrete SSI model and the universal generating function technique, and from this the discrete SSI model under multi-operating conditions is proposed. Finally the proposed model is applied to evaluate the reliability of a transmission component of the decelerator installed in an aeroengine. The reliability of this component during taking-off, cruising and landing phases of an aircraft are calculated, respectively. With this model, a basic method for reliability analysis of the component under complex load condition is provided, and the application range of the conventional SSI model is extended.     

考虑表面粗糙度的干气密封边界膜滑移效应分析

为了研究随机粗糙表面上的滑移流效应,针对符合高斯分布的粗糙度表面,考虑其影响对一阶滑移模型进行修正,建立适用于随机粗糙表面的有效滑移长度模型;建立流体润滑方程,采用数值分析方法对其进行计算,研究粗糙峰间距、粗糙度均方根对滑移流以及螺旋槽干气密封的密封性能的影响。研究结果表明:气体分子平均自由程较小、粗糙峰间距较小或者粗糙度均方根较大的情况下,密封端面更容易发生负滑移现象;在微尺度条件下,负滑移有助于提高密封的端面开启力和运行稳定性;滑移流对气膜刚度的影响最大,气膜承载力次之,泄漏率最小且可以忽略不计。     

Nano-scale roughness effects on hysteresis in micro-scale adhesive contact

A multi-scale model enables investigation of the effect of surface roughness on energy loss in adhesive contact. Fully atomistic simulation of nano-scale indentation predicts size-dependent force–distance trends that are introduced as roughness into a micro-scale finite element model through randomly distributed non-linear hysteretic springs. The multi-scale model predicts that the energy loss, quantified by the hysteresis loop formed by loading and unloading data, increases with increasing indentation depth and surface roughness. This behavior is discussed in terms of an analytical model of a simply connected system.     

基于路面一致激励车桥耦合非平稳随机振动分析

基于路面不平顺一致输入激励,采用虚拟激励法研究车辆变速行驶三维车桥耦合非平稳随机振动响应。首先,将桥梁离散为板-壳实体单元,车辆简化为三维九自由度体系,考虑路面输入激励的多点不相干,将路面不平顺引起的荷载等效为虚拟激励荷载,建立三维车-桥耦合非平稳随机振动模型;然后,运用精细积分格式迭代求解,与Monte-Carlo法计算结果对比验证模型的正确性;最后,以某高速公路梁桥为背景,研究车辆匀加速行驶在B级桥面桥梁各点动响应。结果表明:笔者提出的计算模型及算法正确可行;相同路面激励引起的跨中位移和加速度响应峰值大小取决于瞬时最大车速;车辆变速行驶比匀速行驶具有更宽的共振频率区间,跨中位移和加速度最大值随车速呈现先快后缓的增长趋势。     

Magnetic-Pulse Shaping of Internal Pipe-Blank Surfaces

A finite-element model is proposed for the shaping of internal pipe-blank surfaces by means of the pressure of a magnetic pulse. The roughness of the internal surface after compression onto a mandrel is expressed as a function of the number of loading cycles, the pressure of the pulsed magnetic field, and the initial roughness of the blank.     

Reflections on and Extensions of the Fuller and Tabor Theory of Rough Surface Adhesion

The classical Fuller and Tabor theory of rough surface adhesion is reviewed, and its limited applicability noted. New results using an extension of the JKR adhesion theory to include the effect of van der Waals adhesion forces are described. These show the influence of the Tabor parameter μ: moderate values of µ (µ<100) lead to increased forces during loading, and so are found to predict much lower hysteresis losses. A further extension is to replace the GW surface roughness model used by Fuller and Tabor by a model based on the Nayak model of a surface as a random field, as in the Bush, Gibson and Thomas (BGT) theory of non-adhesive contact. The results are qualitatively the same, and the differences may well be largely due to the necessary redefinition of the adhesion index using the rms profile curvature in place of the asperity radius of curvature used in the GW model. `     

界面因素对机械结合面超声传播的影响

利用超声脉冲回波法可以测量机械结合面间的压力分布,但是其测量精度受界面接触条件的影响,如压力加载历史、表面粗糙度、接触介质等。根据超声在接触界面的传播机制,分析得到接触界面回波能量系数与界面接触压力之间的关系模型;搭建室内超声检测系统,加工不同粗糙度的接触表面,准备不同的润滑介质,对不同接触条件的试件进行压力加载试验,得到结合面回波能量系数随压力变化的曲线。结果表明：首次加载过程所得曲线明显区别于后几次加载,随循环次数的增加,所得曲线逐渐趋于重合;相同接触压力下,接触表面粗糙度越大,回波能量系数越大;相同接触压力下,接触介质的物质属性不同以及与不同粗糙度的表面相互作用时,回波能量系数有所不同。研究表明,载荷加载历史、接触表面形貌及接触介质都是影响超声在接触界面传播的主要因素,研究结果对结合面接触应力分布的准确测量具有参考价值。     

随机粗糙微通道内部流动与传质特性

为了研究微通道壁面随机粗糙度对流体流动和传质特性的影响,采用随机排布准则构建具有典型粗糙元类型的随机粗糙微通道壁面,利用有限元方法分析壁面随机粗糙度对流速、压降、流动阻力和传质性能的影响,并给出粗糙微通道内部Poiseuille数和分子传质扩散的近似变化规律。结果表明,流体在粗糙微通道近壁面区域和主流区的流速差异较大,近壁面区域流动分离现象明显;与光滑微通道相比,粗糙微通道内部各位置的压降和Poiseuille数沿着流动方向呈近似线性增大趋势;微通道壁面粗糙度的存在可以强化流体分子的传质扩散速率,但受粗糙度类型和相对粗糙度的影响较大。     

硅压力传感器的随机振动步进试验定量分析评估

步进应力试验是一项不同于传统可靠性模拟实验的激发式试验技术,其属于可靠性强化试验范畴。在对硅压阻式压力传感器失效模式和失效机理调研的基础上,对其进行随机振动步进试验研究。通过基于传统设备的随机振动步进试验,使产品设计和工艺缺陷得以暴露,提出相关改进措施,并根据应力—强度干涉模型(stress-strength interference model)计算其可靠度。     

Experimental and Numerical Study of Micropitting Initiation in Real Rough Surfaces in a Micro-elastohydrodynamic Lubrication Regime

Micropitting is a form of surface fatigue damage that happens at the surface roughness scale in lubricated contacts in commonly used machine elements, such as gears and bearings. It occurs where the specific film thickness (ratio of smooth surface film thickness to composite surface roughness) is sufficiently low for the contacts to operate in the mixed lubrication regime, where the load is in part carried by direct asperity contacts. Micropitting is currently seen as a greater issue for gear designers than is regular pitting fatigue failure as the latter can be avoided by control of steel cleanliness. This paper describes the results of both theoretical and experimental studies of the onset of micropitting in test disks operated in the mixed lubrication regime. A series of twin disk mixed-lubrication experiments were performed in order to examine the evolution of micropitting damage during repeated cyclic loading of surface roughness asperities as they pass through the contact. Representative measurements of the surfaces used in the experimental work were then evaluated using a numerical model which combines a transient line contact micro-elastohydrodynamic lubrication (micro-EHL) simulation with a calculation of elastic sub-surface stresses. This model generated time-history of stresses within a block of material as it passes through the contact, based on the instantaneous surface contact pressure and traction at each point in the computing mesh at each timestep. This stress time-history was then used within a shear-strain-based fatigue model to calculate the cumulative damage experienced by the surface due to the loading sequence experienced during the experiments. The proposed micro-EHL model results and the experimental study were shown to agree well in terms of predicting the number of loading cycles that are required for the initial micropitting to occur.