| 引用本文: | 姚姝,张辉东,周晓洁,滕军.空间单层网壳结构损伤杆件位置识别试验[J].哈尔滨工业大学学报,2019,51(6):116.DOI:10.11918/j.issn.0367-6234.201809195 |
| YAO Shu,ZHANG Huidong,ZHOU Xiaojie,TENG Jun.Test on member damage location identification of single-layer reticulated shell[J].Journal of Harbin Institute of Technology,2019,51(6):116.DOI:10.11918/j.issn.0367-6234.201809195 |
|
| 本文已被：浏览 1626次 下载 831次 |
码上扫一扫！ |
|
|
| 空间单层网壳结构损伤杆件位置识别试验 |
|
姚姝1,2,张辉东1,2,周晓洁1,2,滕军3
|
|
(1.天津城建大学 土木工程学院,天津300384; 2.天津市土木建筑结构防护与加固重点实验室(天津城建大学),天津300384; 3.哈尔滨工业大学(深圳) 土木与环境工程学院,广东 深圳518055)
|
|
| 摘要: |
| 为解决空间钢结构震后常发生杆件损伤的位置识别问题,设计一个单层网壳结构试验模型,切断单个杆件模拟实际结构的杆件损伤,对该损伤杆件进行位置识别研究.首先,根据空间钢结构自由度多,模态密集等特点,基于小波包能量分析法处理频率密集信号的优势,应用单层钢网壳结构损伤前后两个监测阶段的加速度响应数据,建立抗噪性能和灵敏性较好的损伤杆件位置识别指标向量.再应用有限元软件ANSYS建立17组不同位置杆件损伤的数值分析模型,分别获得17组有限元模型损伤前后的加速度响应数据,通过小波包分析处理后将其损伤指标向量作为损伤识别样本库,再将具有单个杆件损伤试验结构的加速度响应数据处理后得到的损伤识别指标向量与损伤样本库中各杆件损伤位置的指标向量进行聚类分析,识别出实际结构杆件有可能的损伤位置.该方法能够快速识别出实际结构中损伤杆件可能损伤的区域位置,试验结果也验证了方法的可行性和有效性,为后续工程应用研究提供了理论和试验支持. |
| 关键词: 单层网壳 杆件损伤 位置识别 小波包能量分析 聚类分析 |
| DOI：10.11918/j.issn.0367-6234.201809195 |
| 分类号:TU375.2 |
| 文献标识码:A |
| 基金项目:国家自然科学基金青年基金(51508374) |
|
| Test on member damage location identification of single-layer reticulated shell |
|
YAO Shu1,2,ZHANG Huidong1,2,ZHOU Xiaojie1,2,TENG Jun3
|
|
(1.Department of Civil Engineering, Tianjin Chengjian University, Tianjin 300384, China; 2.Engineering and Safety Management Department Key Lab of Urban Security and Disaster Engineering (Tianjin Chengjian University),Tianjin 300384, China; 3.School of Civil and Environmental Engineering,Harbin Institute of Technology(Shenzhen), Shenzhen 518055, Guangdong,China)
|
| Abstract: |
| To identify the damage member location for space steel structures after earthquake, a single-layer reticulated shell structure test model was designed with a single damaged member to study the location identification method. In consideration of the characteristics of spatial steel structure, such as multi-degree of freedom and dense modal, the acceleration responses of two monitoring stages(i.e., before and after structural damage)were used to establish the damage identification index based on wavelet packet energy analysis. The numerical analysis data of 17 different damage locations of members were established as the damage sample base using the finite element software ANSYS. The acceleration response data of the test structure with single member damage were processed and clustered with the damage sample base data to identify possible damage location of the actual structural members. The method can quickly identify the possible damage location of the damaged member in the actual structure and the experimental results validated the feasibility and effectiveness of the proposed method, which provides theoretical and experimental supports for engineering application. |
| Key words: single-layer reticulated shell member damage location identification wavelet packet energy analysis cluster analysis |
