Reliability sensitivity of multi-degree-of-freedom uncertain nonlinear systems with independent failure modes

The reliability sensitivity analysis of uncertain multi-degree-of-freedom nonlinear vibration systems with independent failure modes subjected to random excitation is examined. An earlier version of the statistical fourthmoment method is extended to deal with vector-valued and matrix-valued functions and is developed to determine the first four moments of the system response and state function. Random variables and system derivatives are conveniently arranged into 2D matrices by means of Kronecker algebra. The distribution function of the system state function is approximately determined by the standard normal distribution functions using Edgeworth series technique, and its reliability and reliability sensitivity are presented and discussed.     

基于性能劣化的可靠性预测与系统开发

为了给数控装备可靠性预测提供更丰富、更精确、更及时的信息源,探讨了运用数控装备使用过程中的性能劣化数据对数控装备可靠性进行预测的方法与工具。给出基于性能状态监测的性能劣化建模与可靠性预测原理,并提出一种基于非平稳自回归积分滑动平均模型的数控装备性能劣化建模方法,以监测与分析数控装备关键性能参数的时变特性;提出基于隐马尔科夫链模型的数控装备可靠性预测方法,以揭示数控装备使用过程中的性能状态变化特征及其可靠性变动情况。在此基础上,开发了数控装备可靠性预测原型软件系统。该系统对于提高数控装备利用率、减少数控装备维修费用以及延长数控装备使用寿命等具有重要的意义。     

基于Copula失效相关结构系统的失效概率计算

针对具有相关失效模式的结构系统可靠性分析问题,提出利用Copula计算结构系统的失效概率。Copula函数可以采用多种边缘分布函数来推求联合分布函数,构造随机变量间的相关结构。该方法分别以各失效模式的功能函数值为Copula分析变量,利用MATLAB软件和Monte Carlo模拟法产生的随机抽样数据估计Copula参数,实现各功能函数间的联合概率分布函数的Copula建模,从而可以利用Copula计算结构系统的失效概率。最后用工程算例验证该理论方法的可行性,为具有相关失效模式的结构系统可靠性分析提供新途径。     

一种基于性能退化数据的电子设备缓变故障预报方法

针对电子设备的测试和故障识别提出了一种基于性能退化数据的缓变故障预报方法.重点针对如何体现电子设备功能模块状态与其性能退化数据和故障树之间的内在关系开展建模研究,将难于定量描述的映射关系通过模糊神经网络转化为定性描述的映射关系,结合时间序列分析对性能退化趋势进行评估和判断,为电子设备缓变故障的预报和故障源定位提供了一种便于工程实现的建模方法.以功能模块为基本的故障定位单位,以性能检测数据、可靠性数据、故障树和故障机理等为输入,建立了基于故障树的时间序列神经网络拟合模型,采用时间序列分析对性能退化数据的波动趋势建立分析模型,利用基于故障决策树的模糊神经网络对系统可能故障来源建立判断识别和预报模型,基于非参数密度估计和假设检验方法对预报可信性建立评估模型,从而提高了故障预报的准确性.     

考虑多失效模式的复杂系统运行可靠性评估

为准确评估复杂系统的运行可靠性,针对复杂系统的失效规律和特点,提出考虑多失效模式的运行可靠性评估方法。采用贝叶斯线性模型融合状态监测信息,评估退化水平,建立基于Gamma随机过程的退化失效可靠性评估模型;针对突发失效样本量小的特点,采用贝叶斯方法,建立基于Weibull分布的突发失效可靠性评估模型。以退化失效可靠性评估和突发失效可靠性评估为基础,应用贝叶斯模型平均技术建立复杂系统的运行可靠性评估模型,定量分析退化失效和突发失效对复杂系统运行可靠性的影响,并采用马尔科夫链-蒙特卡洛方法求解模型。以某型航空发动机为例,说明了该方法的应用过程,验证了该方法的有效性。     

Reliability analysis and micromechanics: A coupled approach for composite failure prediction

This work aims at associating two classical approaches for the design of composite materials: first, reliability methods that allow to account for the various uncertainties involved in the composite materials behaviour and lead to a rational estimation of their reliability level; on the other hand, micromechanics that derive macroscopic constitutive laws from micromechanical features. Such approach relies on the introduction of variabilities defined at the microscale and on the investigation of their consequences on the material macroscopic response through an homogenization scheme. Precisely, we propose here a systematic treatment of variability which involves a strong link between micro- and macroscales and provides a more exhaustive analysis of the influence of uncertainties. The paper intends to explain the main steps of such coupling and demonstrate its interests for material engineering, especially for constitutive modelling and composite materials optimization. An application case is developed throughout on the failure of unidirectional carbon fibre-reinforced composites with a comparative analysis between experimental data and simulation results.     

Reliability prediction system based on the failure rate model for electronic components

Although many methodologies for predicting the reliability of electronic components have been developed, their reliability might be subjective according to a particular set of circumstances, and therefore it is not easy to quantify their reliability. Among the reliability prediction methods are the statistical analysis based method, the similarity analysis method based on an external failure rate database, and the method based on the physics-of-failure model. In this study, we developed a system by which the reliability of electronic components can be predicted by creating a system for the statistical analysis method of predicting reliability most easily. The failure rate models that were applied are MILHDBK-217F N2, PRISM, and Telcordia (Bellcore), and these were compared with the general purpose system in order to validate the effectiveness of the developed system. Being able to predict the reliability of electronic components from the stage of design, the system that we have developed is expected to contribute to enhancing the reliability of electronic components.     

Innovation based two-interval overlap test for failure prediction in measurement systems

An approach to predict failure in the measurement system is developed. Proposed approach uses confidence and tolerance intervals for the innovation of Kalman filter and it is based on the criterion of overlapping these intervals whereby it allows the detection of potential failures in the system. The proposed failure prediction method is well suited to realize detection of constant shifting and slowly varied failures, such as distortion in measuring instruments and gyroscope axes shifting in navigation systems. The method is simple and permits to solve the problem of failure prediction in the measurement system under the conditions of its practical operation.     

多元相关性退化系统可靠性模型

针对具有多个相关性能退化过程的复杂系统可靠性评估问题,采用非线性Wiener过程描述系统性能退化过程,并基于多元Copula函数分析系统性能退化过程间的相关关系.在此基础上,建立了系统可靠性模型,并提出了相应的参数估计方法.以三部件系统为例,验证了参数估计方法、相关性退化模型以及可靠性模型的有效性和精确性,分析了相关性...     

Reliability analysis of degradation with a new independent increment process

Degradation test is an important method to assess the reliability of complex systems and highly reliable products. The effectiveness of a degradation model depends strongly on the suitability of the model to describe the process. This paper proposes a new degradation model in which the characteristics of the widely used stochastic process and degradation path models are considered simultaneously. According to the proposed model, closed-form expressions of the performance distribution, failure time distribution and their percentiles, as well as reliability, can be obtained easily. A one-stage procedure is then developed to estimate the model parameters, based on which, estimations of the performance distribution, failure time distribution, and reliability are also presented in the paper. Finally, simulation studies are conducted to validate the proposed method. Results suggest that the method provides precise estimates even for zero-failure cases or an extremely small sample size of approximately five.     

O形圈剪切破坏的可靠性分析

首先利用商业有限元软件ABAQUS建立了O形密封圈的有限元模型,采用非线性有限元方法计算了O形橡胶圈的剪应力的分布,分析了间隙、初始压缩量、密封槽1:7圆角半径、流体压力以及摩擦系数等设计参数对剪应力的影响.最后计算了以剪切破坏为失效模式的O形密封圈的可靠性指标.     

基于竞争威布尔模型的加工中心可靠性评估

为了实现故障数据按故障机制的分类,使用竞争威布尔模型对加工中心可靠性进行评估,通过故障特征属性和故障信息序列概念的定义,以故障信息序列表示故障,利用模糊聚类的方法对故障信息序列按故障机制相似性进行分类;利用自助法和基于灰色预测的改进自助法实现小子样评估中威布尔模型的参数估计;通过实例验证了该模型和方法的可用性。     

Vibration reliability sensitivity analysis of general system with correlation failure modes

The vibration problem of the general system is the main object of research. The material properties and geometry of general system are random parameters because of the manufacturing environment, technical conditions, manufacturing and installation errors, multiphase materials, features and other factors. According to the relation criterion that the difference between the natural frequency and the driving frequency of general systems is not beyond a specific value, the vibration reliability mode and vibration reliability of general systems are defined considering the correlation of the multi-order natural frequency and the random characteristics of structure size and material, and the vibration reliability analysis method for avoiding the resonant is carried out. The second-order joint failure probability is obtained by using the numerical integration method. Based on the reliability design theory and sensitivity analysis method, the vibration reliability sensitivity of the general system with correlation failure modes is extensively discussed and a numerical method for vibration reliability sensitivity design is presented. The variation regularities of vibration reliability sensitivity are obtained and the effects of random parameters on vibration reliability of the general system are studied. The presented method provided the theoretic basis for the reliability design of the general system. A numerical example demonstrated that the proposed method is effective.     

基于小样本预测模型在机械工程上的应用

为预测短期内机械生产实验的数据分布规律,以更好地指导生产和提高工作效率,充分利用计算机模拟研究的优势,基于金属切削工艺加工的部分实验数据,利用线形回归模型、灰色模型和神经网络模型对实验数据进行了分析处理并建立了小样本数据的综合预测模型。研究结果表明小样本预测模型是可行的,它对机械工程各种相关实验数据的短期预测都具有普遍适用性,同时提供了可以快速了解数据在未来分布规律的方法。     

Quantifying chaotic responses of mechanical systems with backlash component

This paper presents a detailed numerical and experimental study of a mechanical system comprising a backlash element that shows chaotic behaviour in a certain excitation range. It aims to quantify the chaotic behaviour of the responses and correlate the quantification parameters to the parameters of the (non-linear) system, in particular the backlash size. The motivation for this investigation is to be able ultimately to identify the parameters of non-linear systems without necessarily being able to ensure periodic behaviour. Application of surrogate data tests is utilised to prove the presence of the chaotic behaviour in the response. The Simple Non-linear Noise Reduction is applied to the resulting data to have a better interpretation of the chaotic in the response.     

Dynamic behaviour of geared systems with backlash due to stick-slip vibratory motion

Stick-slip phenomena occur with low speed rotors in fluid bearings. A typical case is the turning gear mechanism of large turbomachinery which becomes vulnerable to damage. In other cases such as rolling mills or textile machinery the stick-slip phenomenon influences product quality.An analytical investigation of a linear rotor with a complex turning gear system of many degrees of freedom is presented. Gearing backlash was included in the model. The mechanism of backlash was found to be of considerable importance for the appearance of instability. Velocity and damping were the most influential factors on the amplitude of stick-slip motion and instability.In highly unstable areas of system operation, stick-slip is not present owing to the violent nature of the unstable motion. At the onset of instability and during the transients of stable systems, stick-slip is present. Fatigue damage may result from gear backlash, even in stable systems.     

数控车床故障率的可靠性分析

以国产数控车床为例,对该机床故障部位、故障模式和故障原因三个方面进行故障分析。将数控车床划分为10个子系统,根据数据统计各子系统发生故障的概率,找出故障频发的子系统,深入分析故障模式和故障原因,寻找可靠性改进设计的方向。通过危害度分析,摸清数控车床的薄弱环节,找到关键部件,根据实际情况对数控车床进行可靠性分配和改进设计提供依据。     

Machining forces prediction for peripheral milling of low-rigidity component with curved geometry

This study has developed a model in order to predict machining forces for milling low-rigidity component with curved profile. Differing from conventional models used for straight geometries, the new model takes both the deflection of work piece-cutter system and the continuous change of curvature of machined part into account. In the model, a profile with variable curvature is defined and the feed rate per tooth on the profile is derived subsequently. The cutting force is analyzed simulatively by utilizing modified Newton–Raphson iterative algorithm. The simulative results show that the total radial deflection of workpiece–cutter system is the main factor affecting the change of cutting force. Moreover, the feed rate per tooth and its corresponding cutting force changes according to the curvature of the profile. Finally, experiments were carried out to verify the accuracy of the models.     

Reliability-based design optimization of time-dependent systems with stochastic degradation

A major hurdle in the application of reliability-based design optimization (RBDO) to time-dependent systems is the continual interplay between calculating time-variant reliability (to ensure reliability policies are met) and moving the design point to optimize some objective function, such as cost, weight, size and so forth. In most cases the reliability can be obtained readily using so-called fast integration methods. However, this option is not available when certain stochastic processes are invoked to model gradual damage or deterioration. In this case, sampling methods must be used. This paper provides a novel way to obviate this inefficiency. First, a meta-model is built to relate time-variant system reliability to the entire design space (and noise space if required). A design of experiments paradigm and Monte Carlo simulation using the mechanistic model determines the corresponding system reliability accurately. A moving least-squares meta-model relates the data. Then, the optimization process to find the best design point, accesses the meta-model to quickly evaluate objectives and reliability constraints. Case-studies include a parallel Daniel's system and a series servo control system. The meta-model approach is simple, accurate and very fast, suggesting an attractive means for RBDO of time-dependent systems.

     

基于神经网络的电火花线切割机床可靠性预计

针对电火花线切割机床可靠性数据较少、传统可靠性预计方法无法进行可靠性预计或预计误差较大的问题,提出了基于神经网络进行电火花线切割机床可靠性预计的方法。将可靠度与对应的时间序列相结合,采用时间序列法对其进行统计分析,然后应用神经网络向外推测,从而得出电火花线切割机床可靠度的预计结果。通过实例证明,应用该方法对电火花线切割机床可靠度预计结果的最大误差为0．0082,远远高于其它方法的预计精度。