网架结构损伤检测中的数据融合方法研究
发布时间：2018-09-07 08:45
【摘要】：空间网架结构为高次超静定结构,结构形式复杂,可达到较大的跨度,多用于重要的公共建筑中。网架结构由于施工疏忽、使用不当、偶然荷载等因素造成的损伤难以完全避免,一旦结构损伤未被及时发现,则随时可能发生工程事故。此类事故一般都会带来巨大的财产损失和人身危害,因此对空间网架结构损伤检测进行研究尤为重要。针对此工程问题,本文主要进行以下几方面的研究:(1)本文对现有损伤检测方法进行归纳总结,发现空间网架结构的损伤识别技术尚未成熟。由于网架结构节点和杆件众多,其损伤识别多数停留在数值模拟阶段。故本文建立了网架实验模型,对损伤前后的网架结构分别经过应变测试和加速度测试,采取各杆件的应变值和加速度值。再经过模态分析处理,得到实验网架结构损伤前后的前八阶固有频率和对应振型。(2)分别运用基于坐标模态确认准则的损伤识别方法、基于曲率模态差的损伤识别方法和基于模态柔度比变化率的损伤识别方法,对实验结构的振型数据深入分析,得出损伤识别结果。运用基于应变模态的损伤识别方法对实验结构的应变数据进行分析,也得出其损伤识别结果。从这四个损伤判别结果来看,针对本网架结构,各损伤指标均存在或多或少的误判行为。(3)为了提高损伤识别的精度,本文提出把数据融合方法引入到损伤识别领域。对数据融合的基本原理、D-S证据理论的识别框架、信任函数等基本概念和其合成规则等进行了简要的介绍。运用D-S证据理论算法,对上述四种损伤识别结果分别进行信任分配后,再用合成规则进行组合,最终得到更加协同、精确的判断结果。经过三个层次的多次融合,得到对128根杆件发生损伤的信任程度,准确的判断出结构在63号杆位置处发生了损伤。这与实验情况是相符的。本文在对空间网架结构损伤识别的数据融合研究中,针对各种单一损伤识别指标对网架结构存在误判行为,提出把D-S数据融合算法应用到空间网架结构的损伤识别中。其理论意义明确并在实验中得到了验证。事实证明,此方法可以准确的识别出损伤位置,对空间网架结构的损伤识别研究具有明显的理论意义和实用价值。
[Abstract]:The space truss structure is a high order statically indeterminate structure with complex structure form, which can reach a large span, and is mostly used in important public buildings. It is difficult to avoid the damage caused by some factors, such as careless construction, improper use and accidental load, etc. Once the structural damage is not detected in time, engineering accidents may occur at any time. This kind of accident usually brings huge property loss and personal harm, so it is very important to study the damage detection of space truss structure. Aiming at this engineering problem, this paper mainly studies the following aspects: (1) this paper summarizes the existing damage detection methods, and finds that the damage identification technology of space truss structures is not yet mature. Because of the large number of joints and members of the truss structure, the damage identification of the truss structure is mostly in the stage of numerical simulation. Therefore, the experimental model of grid structure is established in this paper. The strain and acceleration of the truss before and after damage are measured, respectively, and the strain and acceleration of each member are taken. After modal analysis, the first eight natural frequencies and corresponding modes before and after damage are obtained. (2) the damage identification method based on coordinate modal confirmation criterion is used respectively. The damage identification method based on curvature modal difference and the damage identification method based on the change rate of modal flexibility ratio are used to analyze the modal data of experimental structures and obtain the damage identification results. The strain data of experimental structures are analyzed by using the damage identification method based on strain mode, and the damage identification results are also obtained. According to the results of these four damage discriminations, each damage index has more or less misjudgment behavior. (3) in order to improve the accuracy of damage identification, the data fusion method is introduced into the field of damage identification. The basic principles of data fusion, such as the recognition framework of D-S evidence theory, the basic concepts of trust function and its composition rules, are briefly introduced. The D-S evidence theory algorithm is used to assign the above four damage identification results respectively and then combine them with the synthesis rules to obtain more cooperative and accurate judgment results. After multiple fusion of three levels, the degree of trust in 128 bars is obtained, and the damage of the structure at the 63 bar position is accurately judged. This is consistent with the experiment. In this paper, in the research of data fusion for damage identification of space truss structures, a D-S data fusion algorithm is applied to the damage identification of space truss structures in view of the misjudgment behavior of each single damage identification index. Its theoretical significance is clear and verified in the experiment. It has been proved that this method can accurately identify the damage location and has obvious theoretical significance and practical value for the damage identification of space truss structures.
【学位授予单位】：西安建筑科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU317;TU356
本文编号：2227776
[Abstract]:The space truss structure is a high order statically indeterminate structure with complex structure form, which can reach a large span, and is mostly used in important public buildings. It is difficult to avoid the damage caused by some factors, such as careless construction, improper use and accidental load, etc. Once the structural damage is not detected in time, engineering accidents may occur at any time. This kind of accident usually brings huge property loss and personal harm, so it is very important to study the damage detection of space truss structure. Aiming at this engineering problem, this paper mainly studies the following aspects: (1) this paper summarizes the existing damage detection methods, and finds that the damage identification technology of space truss structures is not yet mature. Because of the large number of joints and members of the truss structure, the damage identification of the truss structure is mostly in the stage of numerical simulation. Therefore, the experimental model of grid structure is established in this paper. The strain and acceleration of the truss before and after damage are measured, respectively, and the strain and acceleration of each member are taken. After modal analysis, the first eight natural frequencies and corresponding modes before and after damage are obtained. (2) the damage identification method based on coordinate modal confirmation criterion is used respectively. The damage identification method based on curvature modal difference and the damage identification method based on the change rate of modal flexibility ratio are used to analyze the modal data of experimental structures and obtain the damage identification results. The strain data of experimental structures are analyzed by using the damage identification method based on strain mode, and the damage identification results are also obtained. According to the results of these four damage discriminations, each damage index has more or less misjudgment behavior. (3) in order to improve the accuracy of damage identification, the data fusion method is introduced into the field of damage identification. The basic principles of data fusion, such as the recognition framework of D-S evidence theory, the basic concepts of trust function and its composition rules, are briefly introduced. The D-S evidence theory algorithm is used to assign the above four damage identification results respectively and then combine them with the synthesis rules to obtain more cooperative and accurate judgment results. After multiple fusion of three levels, the degree of trust in 128 bars is obtained, and the damage of the structure at the 63 bar position is accurately judged. This is consistent with the experiment. In this paper, in the research of data fusion for damage identification of space truss structures, a D-S data fusion algorithm is applied to the damage identification of space truss structures in view of the misjudgment behavior of each single damage identification index. Its theoretical significance is clear and verified in the experiment. It has been proved that this method can accurately identify the damage location and has obvious theoretical significance and practical value for the damage identification of space truss structures.
【学位授予单位】：西安建筑科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU317;TU356
【引证文献】
相关期刊论文 前1条
1 成浩;张波;朱丽华;;某食堂钢结构网架鉴定与分析[J];陕西理工学院学报(自然科学版);2016年02期
相关硕士学位论文 前1条
1 林健;某螺栓球节点网架厂房坍塌事故分析[D];太原理工大学;2016年
，本文编号：2227776
本文链接：https://www.wllwen.com/jingjilunwen/jianzhujingjilunwen/2227776.html
