基于物理信息及数据融合驱动的复杂街区爆炸荷载快速计算方法

黄阳; 罗定坤; 陈素文

doi:10.11883/bzycj-2025-0238

基于物理信息及数据融合驱动的复杂街区爆炸荷载快速计算方法

doi: 10.11883/bzycj-2025-0238

黄阳^1,,
罗定坤¹,
陈素文^{1, 2, ,}

1.
同济大学土木工程学院，上海 200092
2.
同济大学土木工程防灾减灾全国重点实验室，上海 200092

基金项目: 国家自然科学基金（52578603）；上海市科委“科技创新行动计划”（22dz1202900）；同济大学学科交叉联合攻关重点项目（2023-2-ZD-05）

详细信息

作者简介:
黄　阳（1999－　），男，博士研究生，2210010@tongji.edu.cn

通讯作者:
陈素文（1974－　），女，教授，博士生导师，swchen@tongji.edu.cn

中图分类号: O381
计量
- 文章访问数: 515
- HTML全文浏览量: 63
- PDF下载量: 181
- 被引次数: 0
出版历程
- 收稿日期: 2025-08-08
- 修回日期: 2025-10-10
- 网络出版日期: 2025-10-11

A physics-information and data fusion-driven method for rapid prediction of blast loads in complex urban environments

HUANG Yang^1
,,
LUO Dingkun¹,
CHEN Suwen^{1, 2
, ,}

1.
College of Civil Engineering, Tongji University, Shanghai 200092, China
2.
State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai 200092, China

摘要

摘要: 快速准确地评估复杂街区爆炸荷载对实现高效结构抗爆设计及灾后损伤评估具有重要意义，然而传统经验公式、物理模型及数值模拟方法难以兼顾计算效率与预测精度，现有深度学习爆炸荷载预测模型尚难以用于复杂街区场景。为实现复杂街区爆炸荷载快速准确计算，提出了一种物理信息和数据融合驱动的复杂街区爆炸荷载预测方法，其基本思想是“空间分区、逐步推理”策略，分别针对起爆街道和非起爆街道构建快速网络预测模型，并通过各街道间的边界压力协同工作。两种网络预测模型通过分别引入镜像爆源方法、信号距离场和能量密度因子融合流场关键物理特征，并分别采用3D-UNet网络、2D-UNet联合3D-UNet组成的级联网络作为架构。基于验证后的数值模拟方法生成了两种网络的目标数据，并开展了对应模型训练。模型预测性能的评估结果表明：该方法能够准确预测复杂街区压力场的时空演化过程，在起爆街道和非起爆街道中的流场预测结果与数值模拟结果的相对误差在20%以内，有效描述了流场中指定位置的压力时程。双网络协同方法的推理耗时约为对应数值模拟方法计算时间的2%，单一时刻流场数据存储代价小于对应D3PLOT文件的0.2%，显著降低了计算与数据存储代价。研究为大型复杂街区爆炸荷载快速评估提供了新方法，可为城市建筑抗爆设计和评估提供高效决策支持。
- 复杂城市街区 /
- 爆炸荷载 /
- 镜像爆源法 /
- 能量密度因子 /
- 深度学习
Abstract: Rapid and accurate assessment of blast loads in complex urban blocks is critical for efficient blast-resistant structural design and post-disaster damage evaluation. However, traditional methods, including empirical formulas, physical models, and numerical simulations, struggle to simultaneously achieve high computational efficiency and prediction accuracy. Furthermore, existing deep learning-based blast load prediction models are hard to be applied in complex urban block scenarios. To achieve rapid and accurate assessment of blast loads in complex urban street blocks, a physics-information and data fusion-driven method is proposed. The core idea of the method is a “spatial partitioning and progressive inference” strategy, which involves constructing distinct rapid prediction models for “the detonation street” and “non-detonation streets”. These models then collaborate synergistically via their shared boundary pressures to predict the spatiotemporal evolution of the pressure field across the entire urban block. The two network models incorporate the results from method of images, signed distance fields, and energy density factors to integrate key physical features of the flow field. For the architectures, the two models adopt a 3D-UNet and a cascaded network composed of a 2D-UNet and a 3D-UNet, respectively. The target outputs for both networks were generated using a validated numerical simulation method, which were then used to train the models. Evaluation of the model’s predictive performance demonstrates that the proposed method accurately predicts the spatiotemporal evolution of the pressure field. The relative error between the predicted flow field and numerical simulation results is within 20% in both detonation and non-detonation streets. Moreover, the method effectively captures the pressure-time histories at specified locations. The inference time of the proposed dual-network collaborative method is approximately 2% of the computation time of the corresponding numerical simulation, and the flow field storage cost for a single time step is less than 0.2% of a D3PLOT file, thereby significantly reducing computational and storage costs. The research provides a novel method for the rapid assessment of blast loads in large-scale, complex urban blocks, offering efficient decision-making support for the blast-resistant design and evaluation of urban buildings.
- complex urban environment /
- explosion load /
- method of images /
- energy density factor /
- deep learning

HTML全文

图 1 典型街道及复杂街区的平面示意图

Figure 1. Layout diagrams of typical streets and complex block

下载: 全尺寸图片幻灯片

图 2 双网络协同预测框架

Figure 2. Dual-network collaborative prediction framework

下载: 全尺寸图片幻灯片

图 3 起爆街道预测模型的推理过程

Figure 3. Inference procee of the detonation street prediction model

下载: 全尺寸图片幻灯片

图 4 镜像爆源算法示意图

Figure 4. Diagram of the method of images algorithm

下载: 全尺寸图片幻灯片

图 5 非起爆街道预测模型的推理过程

Figure 5. Inference procee of the non-detonation street prediction model

下载: 全尺寸图片幻灯片

图 6 典型街道能量密度因子f_ECF计算示意图

Figure 6. Schematic diagram of energy concentration factor calculation for typical street configurations

下载: 全尺寸图片幻灯片

图 7 数值建模方法验证

Figure 7. Validation of the numerical modelling method

下载: 全尺寸图片幻灯片

图 8 起爆街道数值模板（单位：m）

Figure 8. Numerical templates for detonation street configurations (unit: m)

下载: 全尺寸图片幻灯片

图 9 非起爆街道数值模板（单位：m）

Figure 9. Numerical templates for non-detonation street configurations (unit: m)

下载: 全尺寸图片幻灯片

图 10 复杂街区爆炸场景示意图（单位：m）

Figure 10. Schematic diagram of the explosion scenario in a complex urban block (unit: m)

下载: 全尺寸图片幻灯片

图 11 不同时刻的压力分布及相对误差

Figure 11. Pressure distribution and relative errors of different times

下载: 全尺寸图片幻灯片

图 12 测点压力时程对比

Figure 12. Pressure-time histories comparison at different gauges

下载: 全尺寸图片幻灯片

表 1 数值模拟与试验测得超压峰值结果对比

Table 1. Comparison of peak overpressures from numerical simulations and the experiment

测点	超压峰值
	试验^[3]/kPa	模拟/kPa		模拟与试验的相误差/%
	试验^[3]/kPa	Fedorova等^[31]	本研究	Fedorova等^[31]	本研究
T11	78.65	46.84	45.96	−40.45	−41.56
T21	82.32	61.23	79.33	−25.62	−3.63

下载: 导出CSV

表 2 起爆街道数据集工况设计

Table 2. Explosion scenario design for detonation streets

街道类型	TNT当量/kg	炸点位置/m	街道类型	TNT当量/kg	炸点位置/m
L型街道	0.070	(0.10, 1.40, 0.08)	十字街道	0.070	(0.10, 1.40, 0.08)
		(0.00, 0.50, 0.08)			(0, 0.50, 0.08)
		(−0.20, 0.70, 0.08)			(−0.20, 0.70, 0.08)
	0.035	(0.10, 1.40, 0.08)		0.035	(0.10, 1.40, 0.08)
		(0.00, 0.50, 0.08)			(0, 0.50, 0.08)
		(−0.20, 0.70, 0.08)			(−0.20, 0.70, 0.08)
	0.019	(0.10, 1.40, 0.08)		0.019	(0.10, 1.40, 0.08)
T型街道	0.070	(0.10, 1.40, 0.08)	T型街道	0.070	(0.10, 1.40, 0.08)
		(0.00, 0.50, 0.08)			(0.00, 0.50, 0.08)
		(−0.20, 0.70, 0.08)			(−0.20, 0.70, 0.08)
		(1.00, 1.50, 0.08)			(1.00, 1.50, 0.08)
		(0.90, 1.40, 0.08)			(0.90, 1.40, 0.08)
	0.019	(0.10, 1.40, 0.08)

下载: 导出CSV

表 3 非起爆街道数据集工况设计

Table 3. Explosion scenario design for non-detonation streets

非起爆街道类型	拼接街道宽度/m	TNT当量/kg	炸点位置/m
L型街道	0.5	0.05	(−0.2, 0.7, 0.08)
	1.0	0.05	(−0.2, 0.7, 0.08)
	0.5	0.07	(0, 0.5, 0.08)
	1.0	0.07	(0, 0.5, 0.08)
T型街道	0.5	0.05	(−0.2, 0.7, 0.08)
	1.0	0.05	(−0.2, 0.7, 0.08)
	0.5	0.07	(0, 0.5, 0.08)
	1.0	0.07	(0, 0.5, 0.08)
十字街道	0.5	0.05	(−0.2, 0.7, 0.08)
	1.0	0.05	(−0.2, 0.7, 0.08)
	0.5	0.07	(0, 0.5, 0.08)
	1.0	0.07	(0, 0.5, 0.08)

下载: 导出CSV

表 4 计算及数据存储代价对比

Table 4. Comparison of computing and data storage costs

方法	计算时间/min	单一时刻流场数据存储/MB
数值模拟	340	386
双网络协同预测方法	7	0.442

下载: 导出CSV

参考文献(34)

[1]	SMITH P D, WHALEN G P, FENG L J, et al. Blast loading on buildings from explosions in city streets [J]. Proceedings of the Institution of Civil Engineers - Structures and Buildings, 2001, 146(1): 47–55. DOI: 10.1680/stbu.2001.146.1.47.
[2]	ROSE T A, SMITH P D. Influence of the principal geometrical parameters of straight city streets on positive and negative phase blast wave impulses [J]. International Journal of Impact Engineering, 2002, 27(4): 359–376. DOI: 10.1016/S0734-743X(01)00060-4.
[3]	SMITH P D, ROSE T A. Blast wave propagation in city streets—an overview [J]. Progress in Structural Engineering and Materials, 2006, 8(1): 16–28. DOI: 10.1002/pse.209.
[4]	FOUCHIER C, LABOUREUR D, YOUINOU L, et al. Experimental investigation of blast wave propagation in an urban environment [J]. Journal of Loss Prevention in the Process Industries, 2017, 49: 248–265. DOI: 10.1016/j.jlp.2017.06.021.
[5]	REMENNIKOV A M. Evaluation of blast loads on buildings in urban environment [M]//JONES N, BREBBIA C A. Structures under Shock and Impact VIII. UK: WIT Press, 2004: 73–82.
[6]	REMENNIKOV A M, ROSE T A. Modelling blast loads on buildings in complex city geometries [J]. Computers & Structures, 2005, 83(27): 2197–2205. DOI: 10.1016/j.compstruc.2005.04.003.
[7]	ZHOU X Q, HAO H. Prediction of airblast loads on structures behind a protective barrier [J]. International Journal of Impact Engineering, 2008, 35(5): 363–375. DOI: 10.1016/j.ijimpeng.2007.03.003.
[8]	US Department of Defense. UFC 3-340-02 Structures to resist the effects of accidental explosions [S]. Washington: The US Department of Army, 2008.
[9]	BAKER W E, COX P A, WESTINE P S, et al. Explosion hazards and evaluation [M]. Amsterdam: Elsevier Scientific Pub. Co. , 1983: 111–209.
[10]	NEEDHAM C E. Blast waves [M]. Cham: Springer, 2018: 363–380.
[11]	CHAN P C, KLEIN H H. A study of blast effects inside an enclosure [J]. Journal of Fluids Engineering, 1994, 116(3): 450–455. DOI: 10.1115/1.2910297.
[12]	杨亚东, 李向东, 王晓鸣. 长方体密闭结构内爆炸冲击波传播与叠加分析模型 [J]. 兵工学报, 2016, 37(8): 1449–1455. DOI: 10.3969/j.issn.1000-1093.2016.08.016. YANG Y D, LI X D, WANG X M. An analytical model for propagation and superposition of internal explosion shockwaves in closed cuboid structure [J]. Acta Armamentarii, 2016, 37(8): 1449–1455. DOI: 10.3969/j.issn.1000-1093.2016.08.016.
[13]	REMENNIKOV A M, MENDIS P A. Prediction of airblast loads in complex environments using artificial neural networks [M]//JONES N, BREBBIA C A. Ninhth International Conference on Structures under Shock and Impact. UK: Wessex Institute of Technology, 2006: 269–278.
[14]	DENNIS A A, RIGBY S E. The direction-encoded neural network: a machine learning approach to rapidly predict blast loading in obstructed environments [J]. International Journal of Protective Structures, 2024, 15(3): 455–483. DOI: 10.1177/20414196231177364.
[15]	杜晓庆, 何益平, 邱涛, 等. 基于PCA-BPNN的桥梁爆炸荷载时程预测 [J]. 爆炸与冲击, 2025, 45(3): 033201. DOI: 10.11883/bzycj-2023-0343. DU X Q, HE Y P, QIU T, et al. Prediction of blast loads on bridge girders based on PCA-BPNN [J]. Explosion and Shock Waves, 2025, 45(3): 033201. DOI: 10.11883/bzycj-2023-0343.
[16]	KANG M A, PARK C H. Prediction of peak pressure by blast wave propagation between buildings using a conditional 3D convolutional neural network [J]. IEEE Access, 2023, 11: 26114–26124. DOI: 10.1109/ACCESS.2023.3257345.
[17]	HUANG Y, ZHU S J, CHEN S W. Deep learning-driven super-resolution reconstruction of two-dimensional explosion pressure fields [J]. Journal of Building Engineering, 2023, 78: 107620. DOI: 10.1016/j.jobe.2023.107620.
[18]	LI Q L, WANG Y, SHAO Y D, et al. A comparative study on the most effective machine learning model for blast loading prediction: from GBDT to Transformer [J]. Engineering Structures, 2023, 276: 115310. DOI: 10.1016/j.engstruct.2022.115310.
[19]	LI Q L, WANG Y, CHEN W S, et al. Machine learning prediction of BLEVE loading with graph neural networks [J]. Reliability Engineering & System Safety, 2024, 241: 109639. DOI: 10.1016/j.ress.2023.109639.
[20]	李般若, 霍璞, 喻君. 基于GNN的爆炸压力时空分布预测模型 [J/OL]. 爆炸与冲击, (2025-06-16)[2025-08-08]. https://link.cnki.net/urlid/51.1148.O3.20250616.1620.032. DOI: 10.11883/bzycj-2024-0503. LI B R, HUO P, YU J. GNN-based predictive model for spatial and temporal distribution of blast overpressure [J/OL]. Explosion and Shock Waves, (2025-06-16)[2025-08-08]. https://link.cnki.net/urlid/51.1148.O3.20250616.1620.032. DOI: 10.11883/bzycj-2024-0503.
[21]	WANG Z Q, PENG J Z, HU J, et al. BlastGraphNet: an intelligent computational method for the precise and rapid prediction of blast loads on complex 3D buildings using graph neural networks [J]. Engineering, 2025, 49: 205–224. DOI: 10.1016/j.eng.2025.03.007.
[22]	HUANG Y, CHEN S W, CHEN X. Pressure–time history prediction for fully and partially confined explosions by combining physical and deep learning models [J]. Engineering Structures, 2025, 339: 120575. DOI: 10.1016/j.engstruct.2025.120575.
[23]	ОРЛЕНКО Л П. 爆炸物理学 [M]. 孙承纬, 译. 北京: 科学出版社, 2011: 101–341.
[24]	HENRYCH. The dynamics of explosion and its use [M]. Amsterdam: Elsevier, 1979: 149–177.
[25]	WU C Q, HAO H. Modeling of simultaneous ground shock and airblast pressure on nearby structures from surface explosions [J]. International Journal of Impact Engineering, 2005, 31(6): 699–717. DOI: 10.1016/j.ijimpeng.2004.03.002.
[26]	LI Y F, CHANG J T, WANG Z A, et al. An efficient deep learning framework to reconstruct the flow field sequences of the supersonic cascade channel [J]. Physics of Fluids, 2021, 33(5): 056106. DOI: 10.1063/5.0048170.
[27]	ÇIÇEK Ö, ABDULKADIR A, LIENKAMP S S, et al. 3D U-Net: learning dense volumetric segmentation from sparse annotation [C]//Proceedings of 19th International Conference on Medical Image Computing and Computer-Assisted Intervention – MICCAI 2016. Athens: Springer, 2016: 424–432. DOI: 10.1007/978-3-319-46723-8_49.
[28]	RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation [C]//Proceedings of 18th International Conference on Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015. Munich: Springer, 2015: 234–241. DOI: 10.1007/978-3-319-24574-4_28.
[29]	SILVESTRINI M, GENOVA B, LEON TRUJILLO F J. Energy concentration factor. A simple concept for the prediction of blast propagation in partially confined geometries [J]. Journal of Loss Prevention in the Process Industries, 2009, 22(4): 449–454. DOI: 10.1016/j.jlp.2009.02.018.
[30]	钟巍, 田宙, 寿列枫. 基于能量密度因子法的复杂环境下爆炸冲击波超压峰值解析计算 [J]. 科学技术与工程, 2021, 21(28): 11947–11954. DOI: 10.3969/j.issn.1671-1815.2021.28.007. ZHONG W, TIAN Z, SHOU L F. Analytical calculation of the blast wave peak overpressure under complex environments based on the energy concentration factor method [J]. Science Technology and Engineering, 2021, 21(28): 11947–11954. DOI: 10.3969/j.issn.1671-1815.2021.28.007.
[31]	FEDOROVA N N, VALGER S A, FEDOROV A V. Simulation of blast action on civil structures using ANSYS Autodyn [J]. AIP Conference Proceedings, 2016, 1770(1): 020016. DOI: 10.1063/1.4963939.
[32]	甘露, 陈力, 宗周红, 等. 近距离爆炸比例爆距的界定标准及荷载模型 [J]. 爆炸与冲击, 2021, 41(6): 064902. DOI: 10.11883/bzycj-2020-0194. GAN L, CHEN L, ZONG Z H, et al. Definition of scaled distance of close-in explosion and blast load calculation model [J]. Explosion and Shock Waves, 2021, 41(6): 064902. DOI: 10.11883/bzycj-2020-0194.
[33]	Federal Emergency Management Agency. Incremental protection for existing commercial buildings from terrorist attack [S]. Washington: Department of Homeland Security, Science and Technology Directorate, 2008.
[34]	KONG X S, ZHOU H, XU J B, et al. Scaling of confined explosion and structural response [J]. Thin-Walled Structures, 2023, 186: 110656. DOI: 10.1016/j.tws.2023.110656.