基于灰色关联理论和遗传算法的悬索桥损伤识别研究

发布时间：2018-03-08 06:06

本文选题：自锚式悬索桥　切入点：损伤识别　出处：《兰州交通大学》2015年硕士论文　论文类型：学位论文

【摘要】：结构工程在服役期内,长期受到静、动态荷载作用,结构材料遭到大气环境的腐蚀和破坏,随着时间的增长必有损伤发生。如若结构损伤不能得以发现并及时修复,必然存在着极大的安全隐患。尤其桥梁结构是公路和铁路的咽喉,一旦发生破坏造成的损伤和影响是极大的。所以及时检测发现桥梁损伤位置及损伤大小,进行维修加固并对桥梁结构工作状态进行安全评估,是当下的热门研究课题,同时也具有非常重要的研究意义和实用意义。结构损伤识别是桥梁结构健康监测的重要组成部分,现阶段根据测量数据的不同将结构损伤检测识别分为两大类:即基于静态测量数据和基于动态测量数据进行损伤识别,利用结构的动静态参数的变化来识别结构的损伤是一个研究热点,相比于动态数据,静态数据易采集且精度较高,对结构损伤较敏感,因此本文主要是基于静态测量数据进行悬索桥的损伤识别。本文在系统了解结构损伤识别方法和现状的基础上,基于灰色关联理论和遗传算法两步法进行损伤识别。通过对自锚式悬索桥损伤识别数值研究证明了该方法有效性。主要研究内容如下:(1)运用Midas有限元软件建立跨湟水河自锚式悬索桥的有限元模型,进行初始平衡状态分析。对于自锚式悬索桥,其易损构件为主梁,其损伤后会对吊杆产生影响,使吊杆内力重新分配。因此在自锚式悬索桥有限元模型中,通过调整不同吊杆的初拉力改变吊杆力模拟由于主梁损伤引起的吊杆内力变化,并提取由此引起的静态数据位移变化值进行损伤识别;(2)基于灰色关联理论知识,运用静态位移曲率置信因子(SDCAC)识别桥梁损伤发生位置。根据损伤前后结构节点位移曲率置信因子大小,判断节点损伤前后的位移曲率关联度大小,进而识别损伤发生的位置。通过对简支梁损伤识别结果得出该方法对结构局部位置损伤定位的有效性。十种吊杆力变化工况下湟水河自锚式悬索桥桥跨方向上梁单元节点SDCAC曲线变化,判断不同的损伤位置,结果表明该方法对于大型复杂的悬索桥也是有效的。(3)基于遗传算法知识,对基本的遗传算法进行改进,采用最优保存策略及自适应交叉和变异概率,提高遗传算法的优化性能。以结构损伤前后应变差最小建立目标函数,对结构损伤程度大小进行识别。在第一步损伤定位的基础上识别悬索桥在不同损伤工况下的损伤程度。悬索桥数值研究结果该方法识别损伤程度大小的有效性。(4)总结本文研究内容,指出悬索桥损伤识所用方法的优势和存在的不足,并对结构损伤识别未来的发展方向做了展望,将动态和静态数据项结合进行结构的损伤识别。
[Abstract]:During the service period, structural engineering is subjected to static and dynamic loads for a long time, structural materials are corroded and destroyed by atmospheric environment, and damage will occur with the increase of time. If the structural damage can not be found and repaired in time, The bridge structure is the throat of the highway and the railway, and the damage and influence caused by the damage is great. Therefore, the damage location and the damage size of the bridge are detected in time. It is a hot research topic to carry out maintenance and reinforcement and evaluate the safety of bridge structure working state. At the same time, it also has very important research significance and practical significance. Structural damage identification is an important part of bridge structure health monitoring. At present, according to the difference of measurement data, structural damage detection and identification is divided into two categories: static measurement data and dynamic measurement data to identify damage. It is a hot topic to identify the damage of structure by using the change of dynamic and static parameters. Compared with the dynamic data, the static data is easy to collect and the accuracy is high, and it is sensitive to the damage of the structure. Therefore, the damage identification of suspension bridge is mainly based on static measurement data. Based on grey correlation theory and genetic algorithm two-step method for damage identification, the effectiveness of the method is proved by numerical research on self-anchored suspension bridge. The main research contents are as follows: 1) using Midas finite element software to establish span. Finite element Model of Self-anchored suspension Bridge in Huangshui River, For the self-anchored suspension bridge, the damage of the main beam of the vulnerable member will affect the suspension rod and make the internal force of the suspension rod redistribution. Therefore, in the finite element model of the self-anchored suspension bridge, the internal force of the suspension rod will be redistributed, so in the finite element model of the self-anchored suspension bridge, the internal force of the suspension rod will be redistributed. By adjusting the initial tension of different suspenders to change the force of the suspenders, simulating the changes of the internal forces of the suspenders caused by the damage of the main beam, and extracting the values of the static data displacement changes caused by the changes, the damage identification is carried out on the basis of the grey correlation theory. The static displacement curvature confidence factor (SDCAC) is used to identify the location of bridge damage. According to the magnitude of displacement curvature confidence factor of structural nodes before and after damage, the correlation degree of displacement curvature before and after damage is determined. Through the result of damage identification of simply supported beam, the effectiveness of this method for location of local location damage of structure is obtained. On the condition of 10 kinds of suspender force changing working conditions, the upper beam of Huangshui River self-anchored suspension bridge in span direction is obtained from the result of damage identification of simply supported beam. Unit node SDCAC curve change, The results show that this method is effective for large and complex suspension bridges. Based on the knowledge of genetic algorithm, the basic genetic algorithm is improved, and the optimal preservation strategy and adaptive crossover and mutation probability are adopted. In order to improve the optimization performance of genetic algorithm, the objective function is established based on the minimum strain difference before and after structural damage. Based on the first step of damage location, the damage degree of suspension bridge under different damage conditions is identified. The numerical results of the suspension bridge show that the method is effective in identifying the damage degree. Summarizing the contents of this paper, This paper points out the advantages and disadvantages of the damage identification methods of suspension bridges, and looks forward to the development of structural damage identification in the future. The dynamic and static data items are combined to identify the structural damage.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U446;U448.25

【参考文献】