基于健康监测的混凝土斜拉桥安全预后方法研究

发布时间：2018-03-27 12:13

本文选题：结构健康监测　切入点：预应力混凝土斜拉桥　出处：《东南大学》2015年硕士论文

【摘要】：桥梁结构健康监测已成为桥梁状态监测与安全评估的重要手段。本文以长深高速(G25)宿淮盐段淮安大桥健康监测为背景,尝试建立预应力混凝土斜拉桥的安全评估、安全预警和安全预后方法,主要工作如下：1.结合淮安大桥的特点,进行淮安大桥健康监测系统整体设计和实桥监测系统的安装调试工作,并对健康监测测试数据进行初步分析。2. 基于环境振动测试结果对淮安大桥有限元模型进行了模型修正,建立能够反映淮安大桥实际状态的基准有限元模型,并进行静动力计算为淮安大桥安全评估和预警建立参数安全阈值。3. 基于不确定层次分析法、群判断、集值统计原理、变权综合理论以及熵值法理论建立淮安大桥安全性评估综合方法,并进行实桥运营状态的安全评估。4. 联合基准有限元模型和支持向量机模型并基于淮安大桥健康监测系统实测数据,构建淮安大桥安全预警体系。5. 采用改进后神经网络的组合模型,并将其运用到混合高斯粒子滤波器的算法中,建立预后的动态模型,对淮安大桥进行安全预后。得到的主要结论如下：(1)基于淮安大桥基准有限元模型及相关规范,建立了淮安大桥结构安全评估限值及安全预警闽值,弥补了现行规范对在线监测项目限值规定之不足。(2)考虑评估过程中的模糊性和不确定性等,建立了淮安大桥结构安全综合评估方法。近六个月的安全评估结果表明其综合评估得分为84.0分,淮安大桥近期安全状况处于二类状态(较好水平)。(3)建立了淮安大桥结构安全蓝色、黄色及红色预警三级预警体系,六个月的监测数据结果表明,各项在线监测项目均有极少量数据超出蓝色预警指标阈值,无数据超出黄色或红色预警指标阂值,可见淮安大桥近期安全状况良好。(4)运用混合高斯粒子滤波器来建立参数动态预测模型,并结合安全评估和安全预警的方法来建立淮安大桥的安全预后。结果表明,模拟预测值与实测值变化趋势基本一致,预测结果与实测的评估结果较为吻合,表明该预后方法可进一步用于大跨度斜拉桥的安全预后。
[Abstract]:Health monitoring of bridge structure has become an important means of bridge state monitoring and safety assessment. Based on the health monitoring of Huai'an Bridge in Suhuai Salt Section, this paper attempts to establish the safety assessment of prestressed concrete cable-stayed bridge. The main work of the methods of safety early warning and safety prognosis is as follows: 1.According to the characteristics of the Huai'an Bridge, the overall design of the health monitoring system of the Huai'an Bridge and the installation and commissioning of the real bridge monitoring system are carried out. Based on the results of environmental vibration test, the finite element model of Huaian Bridge is modified, and the benchmark finite element model which can reflect the actual state of Huai'an Bridge is established. And the static and dynamic calculation for the Huaian Bridge safety assessment and early warning to establish a parameter safety threshold. 3. Based on the uncertain Analytic hierarchy process (AHP), group judgment, set value statistics principle, The theory of variable weight synthesis and the theory of entropy value method are used to establish a comprehensive method for the safety evaluation of Huaian Bridge. The safety assessment of the actual bridge operation state. 4. Combining the benchmark finite element model and the support vector machine model, and based on the measured data of the Huai'an Bridge Health Monitoring system, The safety early warning system of Huai'an Bridge is constructed. The combination model of improved neural network is adopted and applied to the algorithm of mixed Gao Si particle filter to establish the dynamic model of prognosis. The main conclusions are as follows: (1) based on the benchmark finite element model of Huaian Bridge and the relevant codes, the limit value of structural safety assessment and the threshold value of safety warning for Huaian Bridge are established. It makes up for the deficiency of the limit value of online monitoring project in the current code. (2) considering the ambiguity and uncertainty in the evaluation process, A comprehensive assessment method for structural safety of Huaian Bridge is established. The results of safety assessment in the past six months show that the comprehensive assessment score is 84.0 points. The Huai'an Bridge has recently been in a second-class state of safety (at a good level) and has set up a three-stage early warning system for structural safety of the Huai'an Bridge, namely, blue, yellow and red. The results of the monitoring data for six months show that, A very small number of online monitoring items exceeded the blue early warning threshold, and none exceeded the threshold value of yellow or red warning indicators. It can be seen that Huai'an Bridge is in a good safety condition in the near future.) the hybrid Gao Si particle filter is used to establish the parameter dynamic prediction model, and the safety evaluation and early warning method are combined to establish the safety prognosis of the Huai'an Bridge. The results show that, The predicted values are consistent with the measured values, and the predicted results are in good agreement with the measured ones, which indicates that the method can be further applied to the safe prognosis of long-span cable-stayed bridges.
【学位授予单位】：东南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U446;U448.27

【参考文献】