基于径向基神经网络的钢框架地震易损性分析方法
发布时间：2021-12-24 20:12
建筑结构的地震危险性和可靠性评估是一个不确定的标准。钢框架结构地震破坏风险突出。通常,地震易损性分析是从概率的角度进行的,以保持安全性。初始阶段包括标定材料、几何和地震参数的不确定性。地震易损性分析采用“云”方法完成;该方法需要非线性反应时程分析。对于以指数方式增加计算成本的多个时程分析,需要大量的持续时间。经过这样的分析,得到损伤指标的整个过程是严格的。像人工神经网络这样的机器学习技术正在软计算领域出现。以往的研究表明,径向基函数神经网络（RBFNN）能够在足够的数据条件下预测地震损伤。然而,不同结构的神经网络结构之间缺乏比较。因此,为了更快地预测地震易损性曲线,有必要进一步寻找最佳的神经网络及其结构。本文基于机器学习的概念进行以下研究。（1）介绍钢框架结构仿真参数和性能的选择和均匀分布。采用均匀设计方法对多因素多层次实验进行均匀抽样。同时,对钢框架的有限元分析进行了详细地描述。（2）介绍逐步进行云分析和绘制地震易损性曲线的方法。结构损伤指标采用Park-Ang损伤指数。从非线性时程分析中得到的这些数据被用于人工网络（ANN）的训练、测试和验证,以预测易损性曲线。（3）利用径向基函数...
【文章来源】：哈尔滨工业大学黑龙江省 211工程院校 985工程院校
【文章页数】：81 页
【学位级别】：硕士
【文章目录】：
Abstract
中文摘要
Notations
Chapter 1 Introduction
1.1 Background,objective and significance
1.1.1 Background
1.2 Objective and significance
1.3 Source of the topic
1.4 Research methodology
1.5 Development of seismic fragility analysis in structures
1.5.1 Exploration outside the country
1.5.2 Exploration inside the country
1.5.3 Seismic fragility analysis methods and trends
1.6 Application of neural networks in fragility analysis
1.7 Main contents of the paper
Chapter 2 Parameter selection,sampling and model setup
2.1 Introduction
2.2 Parameter and uncertainty range selection
2.3 Uniform design method
2.4 Ground motion selection criteria
2.5 Selection of ground motions
2.6 Model setup in ANSYS workbench
2.7 Summary
Chapter 3 3D-model analysis and fragility curve
3.1 Introduction
3.2 Nonlinear Response History Analysis (NRHA)
3.2.1 Definition
3.2.2 NRHA vs other analysis
3.2.3 Importance of NRHA in engineering practice
3.2.4 Formulation of NRHA
3.2.5 Application of earthquake acceleration data
3.3 Theory and application of damage model
3.3.1 Definition
3.3.2 Different types of damage index and its use
3.3.3 Formulation of damage model
3.4 Fragility curve
3.4.1 Definition
3.4.2 Classification of damage index range
3.4.3 Cloud analysis
3.4.4 Results
3.5 Comparison of Fragility Curves with different parameters
3.6 Problems and challenges
3.6.1 Obstacles faced during the research
3.6.2 Challenges in the research
3.7 Summary
Chapter 4 Fragility prediction using ANN
4.1 Introduction
4.2 Theory of artificial neural network
4.2.1 Definition
4.2.2 Hard computing vs soft computing
4.2.3 Architecture and structure of ANN
4.2.4 Common terms used in ANN
4.2.5 Classification of artificial neural networks
4.3 Design of neural network
4.4 Training and cross-validation
4.5 Prediction of fragility curve
4.6 Problems and challenges
4.7 Summary
Conclusion
References
Acknowledgement
Resume
【参考文献】：
期刊论文
[1]基于地震观测的钢框架结构有限元模拟与易损性分析[J]. 王飞,康现栋,马洁美,Kalkan Erol. 防灾科技学院学报. 2018(04)
[2]基于性态设计的钢筋混凝土结构地震易损性分析[J]. 张进国,王洋,徐龙军. 哈尔滨工程大学学报. 2018(10)
[3]基于性能的钢框架结构地震易损性分析[J]. 李永梅,李玉占,杨博颜. 工程抗震与加固改造. 2017(04)
[4]考虑锈蚀的钢结构地震易损性分析[J]. 郑山锁,田进,韩言召,徐强,孙乐斌. 地震工程学报. 2014(01)
[5]不同弯矩增大系数钢筋混凝土框架结构地震易损性分析[J]. 张耀庭,马超,郭宗明,杜晓菊,刘昌芳. 建筑结构学报. 2014(02)
[6]强柱系数基于结构易损性的抗震性能评估[J]. 周靖,蔡健,方小丹. 西北地震学报. 2007(03)
[7]基于人工神经网络的土工合成材料加筋挡墙临界高度预测模型(英文)[J]. 周建萍,闫澍旺. 岩土工程学报. 2002(06)
本文编号：3551119
【文章来源】：哈尔滨工业大学黑龙江省 211工程院校 985工程院校
【文章页数】：81 页
【学位级别】：硕士
【文章目录】：
Abstract
中文摘要
Notations
Chapter 1 Introduction
1.1 Background,objective and significance
1.1.1 Background
1.2 Objective and significance
1.3 Source of the topic
1.4 Research methodology
1.5 Development of seismic fragility analysis in structures
1.5.1 Exploration outside the country
1.5.2 Exploration inside the country
1.5.3 Seismic fragility analysis methods and trends
1.6 Application of neural networks in fragility analysis
1.7 Main contents of the paper
Chapter 2 Parameter selection,sampling and model setup
2.1 Introduction
2.2 Parameter and uncertainty range selection
2.3 Uniform design method
2.4 Ground motion selection criteria
2.5 Selection of ground motions
2.6 Model setup in ANSYS workbench
2.7 Summary
Chapter 3 3D-model analysis and fragility curve
3.1 Introduction
3.2 Nonlinear Response History Analysis (NRHA)
3.2.1 Definition
3.2.2 NRHA vs other analysis
3.2.3 Importance of NRHA in engineering practice
3.2.4 Formulation of NRHA
3.2.5 Application of earthquake acceleration data
3.3 Theory and application of damage model
3.3.1 Definition
3.3.2 Different types of damage index and its use
3.3.3 Formulation of damage model
3.4 Fragility curve
3.4.1 Definition
3.4.2 Classification of damage index range
3.4.3 Cloud analysis
3.4.4 Results
3.5 Comparison of Fragility Curves with different parameters
3.6 Problems and challenges
3.6.1 Obstacles faced during the research
3.6.2 Challenges in the research
3.7 Summary
Chapter 4 Fragility prediction using ANN
4.1 Introduction
4.2 Theory of artificial neural network
4.2.1 Definition
4.2.2 Hard computing vs soft computing
4.2.3 Architecture and structure of ANN
4.2.4 Common terms used in ANN
4.2.5 Classification of artificial neural networks
4.3 Design of neural network
4.4 Training and cross-validation
4.5 Prediction of fragility curve
4.6 Problems and challenges
4.7 Summary
Conclusion
References
Acknowledgement
Resume
【参考文献】：
期刊论文
[1]基于地震观测的钢框架结构有限元模拟与易损性分析[J]. 王飞,康现栋,马洁美,Kalkan Erol. 防灾科技学院学报. 2018(04)
[2]基于性态设计的钢筋混凝土结构地震易损性分析[J]. 张进国,王洋,徐龙军. 哈尔滨工程大学学报. 2018(10)
[3]基于性能的钢框架结构地震易损性分析[J]. 李永梅,李玉占,杨博颜. 工程抗震与加固改造. 2017(04)
[4]考虑锈蚀的钢结构地震易损性分析[J]. 郑山锁,田进,韩言召,徐强,孙乐斌. 地震工程学报. 2014(01)
[5]不同弯矩增大系数钢筋混凝土框架结构地震易损性分析[J]. 张耀庭,马超,郭宗明,杜晓菊,刘昌芳. 建筑结构学报. 2014(02)
[6]强柱系数基于结构易损性的抗震性能评估[J]. 周靖,蔡健,方小丹. 西北地震学报. 2007(03)
[7]基于人工神经网络的土工合成材料加筋挡墙临界高度预测模型(英文)[J]. 周建萍,闫澍旺. 岩土工程学报. 2002(06)
本文编号：3551119
本文链接：https://www.wllwen.com/jianzhugongchenglunwen/3551119.html
