The use of lifetime functions in the optimization of interventions on existing bridges considering maintenance and failure costs

In the last decade, it became clear that life-cycle cost analysis of existing civil infrastructure must be used to optimally manage the growing number of aging and deteriorating structures. The uncertainties associated with deteriorating structures require the use of probabilistic methods to properly evaluate their lifetime performance. In this paper, the deterioration and the effect of maintenance actions are analyzed considering the performance of existing structures characterized by lifetime functions. These functions allow, in a simple manner, the consideration of the effect of aging on the decrease of the probability of survival of a structure, as well as the effect of maintenance actions. Models for the effects of proactive and reactive preventive maintenance, and essential maintenance actions are presented. Since the probability of failure is different from zero during the entire service life of a deteriorating structure and depends strongly on the maintenance strategy, the cost of failure is included in this analysis. The failure of one component in a structure does not usually lead to failure of the structure and, as a result, the safety of existing structures must be analyzed using a system reliability framework. The optimization consists of minimizing the sum of the cumulative maintenance and expected failure cost during the prescribed time horizon. Two examples of application of the proposed methodology are presented. In the first example, the sum of the maintenance and failure costs of a bridge in Colorado is minimized considering essential maintenance only and a fixed minimum acceptable probability of failure. In the second example, the expected lifetime cost, including maintenance and expected failure costs, of a multi-girder bridge is minimized considering reactive preventive maintenance actions.     

一台220kV电容式电压互感器的故障分析

通过一台电容式电压互感器的故障分析，提出对电容式电压互感器的维护以及预防性试验的一些看法。     

Numerical Cracking and Debonding Analysis of RC Beams Reinforced with FRP Sheet

Bonding a fiber reinforced polymer (FRP) sheet to the tension-side surface of reinforced concrete (RC) structures is often performed to upgrade the flexural capacity and stiffness. Except for upper concrete crushing, FRP sheet reinforcing RC structure may fail in sheet rupture, sheet peeloff failure due to opening of a critical diagonal crack, or concrete cover delamination failure from the sheet end. Accompanying the occurrence of these failure modes, reinforcing effects of the FRP sheet will be lost and load-carrying capacity of the RC structures will be decreased suddenly. This study is devoted to developing a numerical analysis method by using a three-dimensional elasto-plastic finite element method to simulate the load-carrying capacity of RC beams failed in the FRP sheet peeloff mode. Here, the discrete crack approach was employed to consider geometrical discontinuities such as opening of cracks, slipping of rebar, and debonding of the FRP sheet. Comparisons between analytical and experimental results confirm that the proposed numerical analysis method is appropriate for estimating the load-carrying capacity and failure behavior of RC beams flexurally reinforced with a FRP sheet.     

G105钻杆腐蚀失效分析

采用金相分析并结合X射线衍射、扫描电镜和电子探针能谱分析等方法，分析了某油田G105钻杆的腐蚀失效原因。结果表明，因腐蚀产物脱落造成井下堵水眼事故，而钻杆停钻后因未及时清理泥浆而造成钻杆内表面附着泥浆，含有氯离子的泥浆加速了其对钻杆的氧腐蚀。就停钻腐蚀问题提出了预防措施。     

Analysis and Reduction of Ultrasonic Pitting Defects on Hard Disk Surfaces

It has become of paramount importance to be able to perform failure analysis of defects on hard disk surfaces quickly and accurately. In this study, ultrasonic pitting defects on hard disk surfaces were investigated. A new electrical tagging pattern system, which directly translated the channel bits to magnetic transitions during writing of the magnetic pattern around a defect, was used in the failure analysis of magnetic recording disks to mark defective sectors. With the help of the Kerr magneto-effect, the defective sectors were effectively located for further failure analysis. Results of the failure analysis revealed that there was 30% to 40% loss of the recorded magnetic bit signal due to an increase in the gap between a recording head and a hard disk surface with ultrasonic pitting defects. The experiments performed indicated that low ultrasonic power for disk cleaning could reduce the ultrasonic pitting defects significantly. The maximum failure rate due to single-type ultrasonic pitting defects was reduced from 33.4% to 16.7%, while the maximum failure rate due to cluster-type ultrasonic pitting defects was reduced from 11.1% to 5%.     

Lifetime evaluation of concrete structures under sustained post-peak loading

Experimental tests on crack propagation in concrete under constant post-peak loading are simulated using the finite element method and the cohesive crack model, in both Mode I and Mixed-mode conditions. The time-dependent behaviour of concrete in the process zone is due to the interaction and growth of microcracks, a phenomenon which, for high constant load levels, turns out to be predominant over linear viscoelastic creep in the bulk material. In mechanical systems based on this type of material behaviour (creep and strain-softening taking place simultaneously), the initial value problem is non-parabolic, i.e., the error at one time level is affected by the accumulation of errors introduced at earlier time levels. Despite these difficulties, the scatter in numerical failure lifetime vs. load level turns out to be negligible in Mode I conditions and practically acceptable in Mixed-mode conditions. Therefore the time-dependent behaviour of the process zone can be inferred solely from the results of direct tensile tests.     

Preventive maintenance models with random maintenance quality

In real-world environments it is usually difficult to specify the quality of a preventive maintenance (PM) action precisely. This uncertainty makes it problematic to optimise maintenance policy. This problem is tackled in this paper by assuming that the quality of a PM action is a random variable following a probability distribution. Two frequently studied PM models, a failure rate PM model and an age reduction PM model, are investigated. The optimal PM policies are presented and optimised. Numerical examples are also given.     

基于动作推导引擎下的故障检测方法

本文根据文[2]介绍的软件管理者方法，提出了基于动作推导引擎的软件管理者方法。软件管理者单元是一种自动的实时监控软件故障的软件工具，适用于实时软件系统特别是通信类软件系统的故障侦测。它通过监控目标系统的输入和输出，使用获得的输入和目标系统的管理模型推算出对应此输入序列的期望输出值．与目标系统的实际输出做比较．如果实际输出没有在期望输出集中，则管理者单元断定目标系统出现错误。基于动作推导引擎的软件管理者方法，采用了动作推导引擎产生的目标系统管理模型，使该管理模型独立于软件管理者方法。可以实现管理模块与动作推导引擎同步实时更新．而不会导致软件管理者单元的改动。     

基于工作流的电厂缺陷管理系统

通过对电厂缺陷管理流程的分析，利用一种新的方法实现其工作流定义。在工作流定义的基础上，对电厂缺陷管理进行权限控制、消息传递、业务审计及并发控制。系统实现了电厂缺陷管理的自动化闭环控制，并能适用于不同缺陷管理模式的不同电厂。     

几起35kV集合式并联电容器故障分析

集合式并联电容器已广泛用于电力系统。本文介绍了对几种电容器组运行故障原因的发现和分析 ,以及解决的方法。尤其对由于差压保护动作 ,导致投运失败的原因和处理方法作了详细的介绍。文章最后对电力系统的运行保护设定和制造厂的制造提出了建议