【作者】 杨保岑；

【导师】 黄声享；

【作者基本信息】 武汉大学 ， 大地测量学与测量工程， 2010， 博士

【摘要】 大型桥梁结构的施工过程复杂、影响因素众多、控制要求严格,反映桥梁安全状况的结构与形态参数将随着施工进度而发生明显变化。如何评价结构的阶段线形、受力与安全,及时掌握各种不利条件下结构的状态与行为对于桥梁施工和运营期的安全具有重要意义,通过建立桥梁结构形态监测系统对其进行长期连续监测成为保障结构安全的重要手段。本文利用GPS和测量机器人技术设计并实现一套桥梁结构形态监测系统,针对系统实现中的若干关键问题开展研究工作。论文的主要研究内容和成果如下：(1)系统地回顾了结构监测技术的发展历程,总结了其发展现状及取得的成果,在此基础上提出了对于超千米级跨径的大型桥梁施工结构形态监测有待进一步解决的问题。(2)分析了影响大型桥梁施工及运营安全的因素,详细介绍了桥梁结构形态监测的内容和方法,探讨了桥梁结构形态监测系统的总体设计及其测点布置设计的原则。在此基础上,研究开发一套基于GPS和测量机器人的桥梁结构形态监测系统,实现了系统的硬件选型与集成、软件设计与开发、数据传输与处理以及海量信息的存储与管理,该系统具有远程实时、连续、全天候、无人值守自动监测的特点。(3)提出了基于测量机器人的两种监测模式,即定点跟踪法和定期扫测法,研究其对索塔和钢箱梁上所布设监测点的观测方案,并探讨了这两种监测模式的特点和适用性。(4)基于该监测系统自动化连续运行的特点,研究自适应、稳健的数据序列滤波方法。鉴于经验模式分解对非平稳信号的强大处理能力,研究基于经验模式分解的数据自适应滤波消噪方法。(5)深入研究了利用Vondrak滤波处理监测时间序列的方法。针对滤波平滑因子需要人工选取的不足,采用基于Helmert方差分量估计的方法自动确定平滑因子；对于监测数据中的粗差,采用引入一个惩罚函数的方法实现Vondrak滤波的自动化抗差处理。(6)研究卡尔曼滤波用于GPS监测时间序列的实时滤波方法,利用衰减记忆滤波的方法抑制滤波的发散；对于观测数据中可能含有的粗差,采用抗差估计理论中对权函数的迭代计算实现了抗差卡尔曼滤波,这对于自动化监测系统的实时数据处理是有利的。(7)研究利用桥梁结构形态监测系统的监测时间序列数据进行桥梁安全在线评估的模态参数识别方法。由于时域法模态参数识别受观测数据中噪声的影响较为严重,而经验模式分解具有对信号模态和噪声的良好驾驭能力,采用基于经验模式分解的滤波方法处理监测时间序列；对于随机减量技术处理监测时间序列的不足,研究了其改进算法和参数优化设置等关键问题：提出了采用经验模式分解对数据序列进行预处理,利用改进的随机减量技术提取自由响应信号,然后再进行Ibrahim时域法模态参数识别的方法,通过编制程序实现了上述算法,并将其集成在桥梁结构形态监测系统的软件模块中。(8)依托世界最大跨度斜拉桥——苏通大桥的工程建设,结合桥梁结构形态监测系统在其上部结构施工全过程的应用,本文分析了苏通大桥索塔和主梁的形态与结构随施工进度的变化规律,研究了结构在极端施工工况、温度效应和强风荷载下桥梁的动态响应特征,为苏通大桥上部结构施工安全评估和顺利合拢提供了科学的依据。更多还原

【Abstract】 Construction progress of large bridges is complicated with excessive influential factors and the quality control is strict. The structure and morphology parameters that reflect the safety status of the bridge structures will change significantly with the construction progress. The way of which to evaluate the periodic alignments, internal force and its safety and to acquire the state and the response behavior under severe conditions are of great signification for large bridges under construction or after it is open to traffic. Long-term continuous monitoring with bridge structure dynamic monitoring system serves as the best way to insure the structure safety. Under this circumstance, a bridge structure dynamic monitoring system based on GPS and georobot was developed and this paper researched the key problems. The key research contents and main achievements of this article are as follows.(1) With a systematical review of the development courses of structure monitoring techniques and the summarization of its state of the art and achievements, the key problems of dynamic monitoring for large structures or long-span bridges under construction were proposed.(2) The influential factors of safety of large bridges under construction or after it is open to traffic are analyzed and the contents and methods of bridge structure dynamic monitoring were introduced. The general layout and the monitoring points alignments of the bridge structure dynamic monitoring system were researched. A real-time dynamic monitoring system based on GPS and georobot were developed, which was able to operating continuously and automatically. The key problems of the monitoring system such as its hardware design and integration, the software design and development, the data transmission and processing and its data management and storage methods were researched in detail.(3) Two monitoring modes of the georobot based dynamic monitoring system, which are called fixed-point tracking and the specified period scanning respectively, were put forward. Their monitoring schemes for the towers and the cantilever beams under construction were researched. Then the characteristics and the applicabilities were analyzed.(4) Considering the consecutive operation of the monitoring system, adaptive and robust data filtering methods were of great importance. Empirical Mode Decomposition (EMD) based data filtering methods were researched for its being considered an ideal method in many research domains for non-stationary and non-linear data analysis characteristics.(5) Vondrak filter was researched thoroughly for the time series acquired by the monitoring system. Aiming at the shortcoming of its artificial selection of the smoothing factor, a methods based on Helmert variance component estimation was developed, which could estimate an optimal smoothing factor for the input time series. A punishment function was introduced for the realization of robust Vondrak filter, which was able to restrain gross errors automatically during a filtering process.(6) Kalman filter was researched for the time series position data acquired by the monitoring system in real-time and an adaptive memory attenuated filter algorithm was utilized for the divergence restraining of it. In order to eliminate the outliers mixed in the monitoring data series, an iteration algorithm of the weight function based on the robust estimation theory was utilized, which benefited the real-time data processing of the automatic dynamic monitoring system.(7) Online modal parameter identification for bridge safety estimation based on the time series of the monitoring system were researched. As time domain modal parameter identification methods was sensitive to the noise mixed in the time series, the EMD based filter was utilized to preprocess the data for its powerful ability for discrimination of the modal and noise hiding in the data. Considering the shortcomings of the Random Decrement Technique (RDT) for acquiring free decay response signal, some key parameter settings were thoroughly researched and the EMD based filter was employed to improve it and the results showed improved. With all of the key problem researched thoroughly, a method that synthetically utilized the EMD, modified RDT and the Ibrahim time domain method for modal parameter identification was put forward for the first time. Program codes of the above algorithm were developed and integrated into the software module of the structure dynamic monitoring system.(8) Relying on the construction progress of Sutong Bridge, which is the largest cable-stayed bridge on the earth, this article analyzed the structure behavior characteristics of the towers and cantilever beams utilizing the structure dynamic monitoring system. The author also researched the structure dynamics of the bridge under all kinds of disadvantageous loads such as extreme construction conditions, thermal loading effects and strong winds, which provided scientific evidences for the safety assessment and the construction of the main-span closure for Sutong Bridge.更多还原