节点文献

基于桥梁远程监测信息的剩余寿命预测模式研究

Research on Residual Life Prediction Model Based on Bridge Remote Monitoring Information

【作者】 武电坤

【导师】 周建庭;

【作者基本信息】 重庆交通大学 , 桥梁与隧道工程, 2009, 硕士

【摘要】 桥梁在服役期间,环境侵蚀、材料老化和荷载的长期效应、疲劳效应与突变效应等不利因素的耦合作用将不可避免地导致结构和系统的损伤积累及抗力衰减,其剩余使用寿命经受着严峻的考验。随着桥梁健康监测技术的发展,测试信息的准确性相应提高;同时时变可靠度理论也不断得到完善,极大地提高了预测评估的可靠性。而现有的结构使用寿命预测方法都不能系统地、真实地对桥梁做出可靠的预测,本课题旨在从研究桥梁实时监测信息出发,以时间序列理论和可靠度理论为基础,建立结构抗力和所受荷载效应的数学统计模式,进而达到预测桥梁整体结构的剩余使用寿命目的。首先比较常用结构使用寿命预测方法的优缺点,重点分析基于监测信息的寿命预测特点及优势;利用规范和有限元理论,结合监测手段提出确定桥梁初始状态的三种方法。影响桥梁可靠性剩余寿命预测的因素大体分为结构抗力和荷载效应两类,因此将它们构成整个研究的主题。通过收集国内外对结构抗力时变方面研究的资料,系统总结了诱发桥梁抗力衰变的各个影响因素,并对其影响机理进行深入剖析;借用材料力学和结构动力学的基本原理,推导出反映结构抗力衰变的特征因子;时频分析理论揭示了监测信息与这些特征因子的相互关联。依据数学手段和各特征因子的特性,提出特征因子反映抗力的方法;小波多尺度分析和数学拟合方法的思路解决了监测信息中活载效应分离的问题,为修正现有荷载模型打下基础。最后,应用时间序列理论建立起基于监测信息的结构抗力及荷载效应的预测性时变模式,构建出可靠性结构寿命预测的框架流程模式。以实桥监测信息为依托,分析了该桥抗力及活载效应的动态趋势,给出挠度监测信息的ARMA预测函数表达式,总结、验证了寿命预测模型的正确性。该模型融入信息、数学和力学等多方面知识,它将具有广泛的工程应用前景,及重大的社会效益和经济效益。

【Abstract】 During service of bridge, coupling effects of unfavorable factors such as environmental erosion, material aging, long-tem effect of load, fatigue effect and mutagenic effect lead to damage accumulation and resistance attenuation in the structure and system inevitably, and residual service life of bridge endure severe test. With the development of bridge health monitoring and improvement in accuracy of test information, at the same time, the time-varying reliability theory is consummated continuously; the above improve reliability of prediction and evaluation mostly. Because existing prediction method for structure service life can’t make reliable prediction accurately and systematically, this task form studying on real-time monitoring information of bridge and based on time series theory and reliability theory establish mathematical statistical model between structural resistance and load effect, in order to predicting residual service life of bridge’s whole structure.Firstly, advantage and disadvantage of usual prediction method for structure service life are compared, the characteristic and advantage of life prediction based on monitoring information is analyzed; the three measures of confirming bridge’s initial state are proposed using code, finite element theory and monitoring means. The factors of affecting prediction for bridge reliable residual life are consist of structural resistance and load effect, so the topic of this paper is composed form these. Through the materials of studying on prediction measures of the structure service life overseas are collected , kinds of influencing factors of inducing bridge’s resistance decay are summarized, and influence mechanisms are analyzed deeply. Using the basic theories of material mechanics and structure dynamics, characteristic factors of reflecting structure resistance decay are deduced. The theory time-frequency analysis reveals the interdependence between monitoring information and these characteristic factors. Based on the measure of mathematics and the identify of these characteristic factors, the way that these characteristic factors reflect structure resistance is lodged. Analysis character of wavelet multi-scale and mathematical fitting method provides thoughts for question of live load effects separation in monitoring information, the above offer a correction method for existing load method.Finally, using time series theory, structure resistance based on monitoring information and time-varying mode of load effect are established, the processes establish frame diagram of life prediction of the reliability structure.Based on monitoring information of actual bridges, the dynamic trend of the structure resistance and the effect of live load are analyzed, the function expression which is ARMA prediction of deflection is proposed, and accuracy of life prediction model is summarized and verified.Various knowledge such as information, mathematics and mechanics are crossed in this model. The thought have a wide range of engineering applications, and significant social and economic benefits.

节点文献中: