文章摘要
苏哲,王晓玲,余佳,王佳俊,余红玲,张君.数字孪生框架下土石坝地震动力响应时空分析系统研究[J].水利学报,2025,56(11):1454-1468
数字孪生框架下土石坝地震动力响应时空分析系统研究
Study on spatiotemporal analysis system of earth-rockfill dam seismic dynamic response under the digital twin framework
投稿时间:2025-01-08  修订日期:2025-11-19
DOI:10.13243/j.cnki.slxb.20250007
中文关键词: 大坝地震动力响应分析  数字孪生  材料参数非均质性  时空全域代理模型  三维物理场可视化
英文关键词: dam seismic dynamic response analysis  digital twin  material parameters heterogeneity  spatiotemporal global surrogate model  3D physical field visualization
基金项目:国家自然科学基金重大项目(52494970)
作者单位E-mail
苏哲 天津大学 水利工程智能建设与运维全国重点实验室, 天津 300072  
王晓玲 天津大学 水利工程智能建设与运维全国重点实验室, 天津 300072 wangxl@tju.edu.cn 
余佳 天津大学 水利工程智能建设与运维全国重点实验室, 天津 300072  
王佳俊 天津大学 水利工程智能建设与运维全国重点实验室, 天津 300072  
余红玲 中国农业大学 水利与土木工程学院, 北京 100083  
张君 天津大学 水利工程智能建设与运维全国重点实验室, 天津 300072  
摘要点击次数: 29
全文下载次数: 17
中文摘要:
      针对现有大坝地震动力响应分析系统缺乏考虑施工质量对坝料物理力学参数的影响、难以快速分析和可视化动力响应的时域和空间分布特征的问题,结合数字孪生的虚实映射、快速分析等优势,提出数字孪生框架下大坝地震动力响应时空分析系统构建方法。在系统数据底板方面,通过参数化建模、施工质量智能监控、虚拟传感器等方法,构建包含高保真几何模型、坝体材料参数和地震动信息的系统数据底板;其中,采用施工质量影响下基于迁移学习和ResNet的大坝材料参数非均质性分析模型,实现了全坝材料参数空间分布的准确分析。在专业模型方面,提出基于pix2pixHD的大坝地震响应分析时空全域代理模型,通过多层生成对抗神经网络学习大坝地震响应数据的时域分布特征,并基于局部增强器提升分析结果中空间分布特征的真实性,克服现有代理模型仅能预测少数点位,无法表征整个大坝地震动力响应时空分布特征的不足。在系统可视化方面,提出基于CUDA加速三维物理场计算的大坝地震动力响应可视化方法,解决现有研究多采用预制动画进行地震过程可视化的问题。最后,以西南某土石坝工程为例,研发了数字孪生框架下的大坝地震分析系统,仅需148 ms即可快速分析和可视化整个大坝的地震动力响应全过程,为大坝地震安全评估和决策提供有力的支持。
英文摘要:
      To address the shortcomings of current dam seismic dynamic response analysis systems,which often fail to consider the impact of construction quality on the physical and mechanical parameters of dam materials and struggle with rapid analysis and visualization of the temporal and spatial distribution of dynamic responses,this study proposed a method for constructing a spatiotemporal analysis system for dam seismic dynamic response under the digital twin framework. This approach leveraged the strengths of digital twins,such as virtual-real mapping and rapid analysis. The system's data foundation integrated high-fidelity geometric models,material parameters of the dam body,and seismic motion information through parameterized modeling,intelligent monitoring of construction quality,and virtual sensors. A novel heterogeneity analysis model for dam material parameters based on transfer learning and ResNet was introduced to accurately analyze the spatial distribution of material parameters across the entire dam, considering the effects of construction quality. For specialized modeling,a pix2pixHD-based surrogate model for spatiotemporal analysis of dam seismic response is developed. This model used multi-layer generative adversarial networks to learn the temporal distribution characteristics of seismic response data and local enhancers to improve the authenticity of spatial distribution features in the analysis results,thereby overcoming the limitations of existing surrogate models that can only predict a few points and fail to characterize the overall spatiotemporal distribution of seismic dynamic responses. For system visualization,a CUDA-accelerated 3D physical field calculation method was proposed for visualizing dam seismic dynamic responses,addressing the issue of using pre-made animations for earthquake process visualization in current research. Using an earth-rockfill dam project in southwestern China as case study,a dam seismic analysis system was developed within the digital twin framework. The system achieved rapid analysis and visualization of the entire dam's seismic dynamic response in just 148 ms,providing strong support for dam seismic safety assessment and decision-making.
查看全文   查看/发表评论  下载PDF阅读器
关闭