| 基于神经网络的结构震后快速损伤评估 |
| |
| 引用本文： | 杨耀鑫,,杨永强,,杨游,公茂盛,.基于神经网络的结构震后快速损伤评估[J].世界地震工程,2023,39(1):049-58. |
| |
| 作者姓名： | 杨耀鑫 杨永强 杨游 公茂盛 |
| |
| 作者单位： | 1.中国地震局工程力学研究所地震工程与工程振动重点实验室,黑龙江哈尔滨150080;2.地震灾害防治应急管理部重点实验室,黑龙江哈尔滨150080;3.中国市政工程中南设计研究总院有限公司,湖北武汉430010 |
| |
| 基金项目： | 国家自然科学基金资助项目(52078472)；;国家重点研发计划资助(2017YFC1500602)； |
| |
| 摘 要： | 为了利用结构地震响应观测数据在震后对结构进行损伤快速评估，本文提出了基于BP传播神经网络多参数预测震后结构损伤程度的方法。本文设计了9个不同设防烈度和层数的钢筋混凝土框架结构，利用OpenSees有限元软件进行了非线性时程分析，并用损伤指数量化了结构损伤程度。利用有限元模拟结果，创建了神经网络的数据集，训练神经网络建立了结构参数与结构损伤指数之间的映射，对比了不同参数组合预测结构损伤水平的能力，提出了最优参数组合。结果表明：此方法预测结构损伤指数准确度高，耗时短，可为建筑工程震后损伤快速评估提供支撑。
|
| 关 键 词： | 结构地震响应 钢筋混凝土框架结构 神经网络 结构损伤 快速评估 |
Rapid assessment of structural damage after earthquake based on neural network |
| |
| YANG Yaoxin,' target="_blank" rel="external">,YANG Yongqiang,' target="_blank" rel="external">,YANG You,GONG Maosheng,' target="_blank" rel="external">.Rapid assessment of structural damage after earthquake based on neural network[J].World Information On Earthquake Engineering,2023,39(1):049-58. |
| |
| Authors: | YANG Yaoxin " target="_blank">' target="_blank" rel="external"> YANG Yongqiang " target="_blank">' target="_blank" rel="external"> YANG You GONG Maosheng " target="_blank">' target="_blank" rel="external"> |
| |
| Affiliation: | 1. Key Laboratory of Earthquake Engineering and Engineering Vibration, Institute of Engineering Mechanics, China Earthquake Administration,Harbin 150080, China; 2. Key Laboratory of Earthquake Disaster Mitigation, Ministry of Emergency Management, Harbin 150080, China; 3. Central and Southern China Municipal Engineering Design and Research Institute Co., Ltd., Wuhan 430010, China |
| |
| Abstract: | In order to use the structural seismic response observation data for rapid damage assessment of structures, this paper proposes a method based on BP neural network for multi-parameter prediction of the post-earthquake structural damage degree. In this paper, nine reinforced concrete frame structures with different fortification intensities and number of stories were designed, and nonlinear dynamic time history analysis was performed by OpenSees, and the damage index was used to quantify the degree of structural damage. Using the finite element simulation results, a dataset of neural network was created, and the mapping between structural parameters and structural damage indices was established by training the neural network, comparing the ability of different parameter combinations to predict the structural damage level, and proposing the optimal parameter combination. The results show that this method is highly accurate and time-consuming, and can provide support for rapid assessment of post-earthquake damage in construction projects. |
| |
| Keywords: | structural seismic response RC frame structures neural network structural damage rapid assessment |
|
| 点击此处可从《世界地震工程》浏览原始摘要信息 |
|
点击此处可从《世界地震工程》下载全文 |
|
