车辆激励作用下斜拉桥的模态分析与桥梁评估

发布时间：2018-04-28 19:57

本文选题：车辆激励 + 模态分析　；参考：《河北建筑工程学院》2017年硕士论文

【摘要】：随着我国综合国力的提升和一带一路战略的提出,无疑促进了交通事业进一步发展。其中斜拉桥作为重点交通工程的枢纽,建成后的维护和运营越来越显得重要。为了监控桥梁的服役状态,不惜花重金修建健康监测系统。由于环境激励类型多,结构复杂,非线性强,测量噪声种类繁多,所以健康监测系统响应信息量大,数据冗余。结合模态参数求解难度高,误差大的特点。现摒弃传统工作模态分析,采用可控性较强的车辆激励,进行模态分析。研究拟从力学角度出发,优化三轴重型车辆模型,建立车桥耦合数值模型。然后求得单点激励振动方程的理论解,通过数值分析和理论分析来研究桥梁振动响应规律,并找到最贴合实际的车辆激励方式。针对斜拉桥监测信号高噪声的特点。通过小波阈值的方式,对信息进行消噪处理,通过功率谱分析,确定最佳的阈值函数和处理方法。然后采用经验极点对称模态分解法对阈值后的信号进行模态分解,得到若干分解模态。在此基础上采用时域幅值分析法,求得时域幅值参数,分析了损伤结构响应对时域幅值参数的影响。对于斜拉桥这种大型结构,材料非线性和几何非线性对模态分析结果的影响不可忽略。建立了一种以复合实体、膜和梁单元来模拟斜拉桥的方法,其中考虑阻尼器、限位器和预应力。在研究中应用模态子结构技术,来分批计算桥梁的附属结构,解决现在计算机计算能力不足的问题。在分析时,使用初始状态法,考虑小变形假设和大变形假设,解决几何非线性问题。数值分析应用了混凝土损伤本构模型和理想弹塑性本构模型,以此解决材料非线性问题。并且通过分析验证了有限元模型的正确性。采用了小波阈值的方法对振动波进行消噪处理,并通过ESMD方法对消噪后的振动波进行模态分解。在Elman网络的基础上,构建了层次神经网络,并以福彩3D的数据验证了其精确性。在此基础上对桥梁的损伤进行了评估,获得良好的效果。
[Abstract]:With the promotion of China's comprehensive national strength and Belt and Road's strategy, the further development of the transportation industry has undoubtedly been promoted. The cable-stayed bridge as a key transportation project hub, after completion of maintenance and operation is becoming more and more important. In order to monitor the service status of the bridge, the health monitoring system is built at great expense. Because of many kinds of environmental excitation, complex structure, strong nonlinearity and various kinds of measurement noise, the health monitoring system has a large amount of response information and data redundancy. Combining with the characteristics of high difficulty and large error in solving modal parameters. In this paper, the traditional mode analysis is abandoned and the vehicle excitation with strong controllability is used to carry out modal analysis. From the point of view of mechanics, the model of three axle heavy duty vehicle is optimized and the coupling numerical model of vehicle and bridge is established. Then the theoretical solution of the vibration equation of single point excitation is obtained. The vibration response law of bridge is studied by numerical analysis and theoretical analysis, and the most suitable vehicle excitation mode is found. Aiming at the characteristics of high noise monitoring signal of cable-stayed bridge. The best threshold function and processing method are determined by power spectrum analysis. Then the empirical pole symmetric mode decomposition method is used to decompose the signal after threshold, and some decomposition modes are obtained. On the basis of this, the time-domain amplitude parameters are obtained by using the time-domain amplitude analysis method, and the influence of the damage structure response on the time-domain amplitude parameters is analyzed. For the cable-stayed bridge, the influence of material nonlinearity and geometric nonlinearity on the modal analysis results can not be ignored. A method of simulating cable-stayed bridge with composite entity, membrane and beam element is established, in which dampers, limiter and prestress are considered. The modal substructure technique is used to calculate the attached structure of the bridge in batches to solve the problem of insufficient computer computing capacity. In the analysis, the initial state method is used to solve the geometric nonlinear problem with the assumption of small deformation and large deformation. The concrete damage constitutive model and the ideal elastoplastic constitutive model are applied in the numerical analysis to solve the material nonlinearity problem. The correctness of the finite element model is verified by analysis. The wavelet threshold method is used to de-noising the vibration wave and the ESMD method is used to decompose the vibration wave. On the basis of Elman network, a hierarchical neural network is constructed, and its accuracy is verified by 3D data of welfare lottery. On this basis, the damage of the bridge is evaluated and good results are obtained.
【学位授予单位】：河北建筑工程学院
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U448.27

【参考文献】