定时预防维修方针和策略的数字仿真分析

传统上对于装备的定时预防维修有两种方针：更新型方针和修复型方针，针对这两种维修方针，在维修费用策略、系统可用度策略基础上，从效能／费用观点出发提出效费比维修策略，进而对一个典型的威布尔故障分布实例进行数字仿真。结果表明，效费比维修策略在达到最佳效费比的同时，能够显著提高系统可靠性，并已在某型武器维修实践中得到应用。     

考虑颤振效应的超大型集装箱船极限强度

砰击颤振会威胁船体的总纵强度,本文先依据海况对响应贡献率最大的原则确定计算海况,然后计算得到一艘超大型集装箱船的波浪载荷时历,并采用Weibull、Gumbel以及GEV(Generalized Extreme Value distribution)分布拟合得到载荷短期极值沿船长的分布且校核了目标船的极限强度。本文通过短期极值Weibull、Gumbel分布拟合的结果与载荷的规范计算结果对比,发现有必要在考虑砰击颤振效应下利用波浪载荷直接预报结果校核目标船结构强度。     

Application of Small Sample Analysis in Life Estimation of Aeroengine Components

The samples of fatigue life tests for aeroengine components are usually less than 5, so the evaluation of these samples belongs to small sample analysis. The Weibull distribution is known to describe the life data accurately, and the Weibayes method (developed from Bayesian method) expands on the experiential data in the small sample analysis of fatigue life in aeroengine. Based on the Weibull analysis, a program was developed to improve the efficiency of the reliability analysis for aeroengine compgnents. This program has complete cycle fatigue life was evaluated by this program. From the results, the following conclusions were drawn: (a) that this program could be used for the engineering applications, and (b) while a lack of former test data lowered the validity of evaluation results, the Weibayes method ensured the results of small sample analysis did not deviate from the truth.     

飞机起飞和着陆所需滑跑距离计算方法中参数值的试验分析

目前,计算飞机滑行距离的理论公式需要许多参数值,这些参数值通常采用线性内插法和随机取样法得到,但综合修正系数（IMC）却要基于飞机的滑跑长度列线图。由于公式中的IMC取决于试验数据,用激光测距仪和经纬仪对飞机的滑跑距离进行了220次测试。试验分析结果表明:飞机起飞时的IMC介于1.000～1.099之间,而着陆时的IMC介于1.468～1.699之间。由此可见,飞机着陆滑跑长度的离散性较起飞时大。     

基于统计损伤理论的沥青混合料的蠕变模型

将修正的Burgers模型看成是Van Der Pool模型与一个非线性黏壶串联而成,认为材料损伤演化过程只是导致模型中串联黏壶的黏度降低,其他3个元件并没有受到损伤;用Weibull函数来描述沥青混合料内部缺陷的分布,从统计学的角度出发建立了损伤演化方程,将损伤引入修正的Burgers模型的非线性黏壶,建立了沥青混合料的黏弹性损伤模型;给出了蠕变应变、蠕变速度和蠕变加速度的解析表达式,证明了该模型能很好地反映沥青混合料三阶段的蠕变特性;最后通过两个试验算例验证了模型的准确性和适用性.研究结果表明,只考虑串联黏壶的损伤不仅较以往的蠕变损伤模型更加简单,而且能很好地反映实验结果,最重要的是这种处理有了更合理的理论根据.     

基于B/S架构态势原型系统实现及性能优化

C/S（客户机-服务器）模式软件存在客户端资源消耗大、部署繁杂、升级困难等问题。论文设计了一种基于Java平台的B/S（浏览器-服务器）架构态势原型系统,解决了体系结构、信息规范、网络通信及态势显示等关键技术,实现了实时目标信息接入、图上显示及综合查询等功能。论文还讨论了Java平台性能优化技术,优化后的原型系统运行结果良好,在指挥控制系统领域具有应用价值。     

动车组运用维修间隔优化方法的研究

介绍现阶段航空、轨道交通领域维修间隔研究的主要方法,结合动车组维修的特点,以动车组运用维修时间间隔为优化目标,针对现行"计划预防修"体制下的"过剩维修"和"维修不足"并存的现象,基于动车组运用修故障数据,采用生存分析与可靠性理论来确定部件合理的维修间隔。关键步骤主要涵盖部件可靠度曲线的威布尔分布的拟合、利用最大似然法对两参数威布尔分布的参数进行估计以及对拟合出的可靠度函数利用条件概率确定维修间隔。并以CRH380B(L)型动车组轮对空心车轴为例,对文中介绍的方法进行验证。结果表明:对于状态不会因维修活动改变的动车组部件,其维修间隔可根据部件的可靠性确定,基于故障数据的理论计算结果与动车组维修实际吻合度较高,可为动车组维修项目周期的优化提供参考。     

Crash analysis at intersections in the CBD: A survival analysis model

Enhancing the safety level of urban roads especially in CBDs is paramount. Due to a large number of intersections in what is usually a grid road system in the CBDs, we investigate crashes occurring in and around an intersection. The question of interest in this study is: does the nature of crashes at intersections differ from those of the roads at midblock? Stated more precisely, considering the intersection as a reference point, does the distance to the reference point (i.e. midblock locations on the roads) correlate with different types of crashes compared to that of the intersection? A right answer can lead traffic engineers and safety auditors to propose different safety measures at intersections and the midblock locations. As a pilot study, we collected the last 9 years crash data of the CBD of Melbourne, Australia. For the first time, we employ Survival Analysis models -including Exponential, Weibull, and Log-logistic- to investigate a space-dependent phenomenon (i.e. accidents at proximity to the intersection). Of the outcome, highlights are: (i) police presence at busy intersections during busy night outs and weekends highly improves the pedestrian safety (ii) raised crossings at midblock locations lower likelihood of crashes of pedestrians as well as cars, (iii) lighting conditions at intersections must be watched and kept at a high level. (iv) Severity, likelihood, and location have no known association with the level of congestion. In other words, safety is first, always and everywhere. The results can be of interest to traffic authorities and policy makers in reinforcing traffic calming measures in the cities. The codes developed in this study are made available to the research community to be used in further studies.     

随机模拟在干散货船运输效益分析中的应用

分析影响干散货船运输效益的两个主要随机变量:非生产停泊时间、燃油价格。通过统计数据,指出它们的分布特点,并通过假设检验,得到非生产停泊时间符合爱尔朗分布,燃油价格符合威布尔分布的结论。在此基础上,将随机模拟方法应用于10000t级江海直达干散货船的运输效益分析中,得到该船在北仑港到武汉港航线上符合实际情况的必要运费率的累积概率分布值。     

沥青混凝土长期耐久性能在统计学上的分布规律研究

已有研究通常采用正态分布来描述沥青混凝土的疲劳耐久性,但是由于材料本身以及试验环境条件的变异性,试验所得的疲劳耐久性能具有很大的离散性。依据大量室内试验并优选代表性强的数据源,利用双参数Weibull分布理论验证沥青混凝土材料的疲劳耐久性能在统计学上的分布规律,并回归得到了不同失效概率条件下的疲劳寿命方程。结果表明:Weibull分布理论能够较好地描述多因素复杂条件下沥青混凝土的疲劳耐久性分布特征。