Page 22 - 《爆炸与冲击》2025年第5期

第 45 卷 第 5 期 爆 炸 与 冲 击 Vol. 45, No. 5
2025 年 5 月 EXPLOSION AND SHOCK WAVES May, 2025

DOI：10.11883/bzycj-2024-0099

端到端机器学习代理模型构建及其

在爆轰驱动问题中的应用 *

柏劲松，刘 洋，陈 翰，钟 敏

（中国工程物理研究院流体物理研究所冲击波物理与爆轰物理全国重点实验室，四川 绵阳 621999）

摘要： 人工智能/机器学习方法能够发现数据中隐藏的物理规律，构建状态参数与动态结果之间端到端的代理模
型，可高效解决强耦合、非线性、多物理等复杂工程问题。在高度非线性的爆炸与冲击动力学领域，选择了一个经典
的爆轰驱动问题作为研究对象，以数值模拟结果作为机器学习代理模型的训练数据，将正向模拟与逆向设计有机结合
起来，基于深度神经网络技术，构建了特征位置速度剖面、材料动态变形与工程因素之间端到端的代理模型，给出了
代理模型的计算精确度，验证了代理模型从速度剖面反演工程因素的能力。结果表明：端到端代理模型具有较高的预
测能力，其预测的速度剖面与工程因素估计的相对误差均小于 1%，可用于高度非线性的爆炸与冲击动力学问题的快
速设计、高精度预测和敏捷迭代。
关键词： 计算爆炸力学；爆轰驱动；人工智能；机器学习；端到端代理模型；深度神经网络
中图分类号： O389 国标学科代码： 13035 文献标志码： A

Construction of end-to-end machine learning surrogate model
and its application in detonation driving problem

BAI Jingsong, LIU Yang, CHEN Han, ZHONG Min
（National Key Laboratory of Shock Wave and Detonation Physics, Institute of Fluid Physics,
China Academy of Engineering Physics, Mianyang 621999, Sichuan, China）

Abstract: Artificial intelligence/machine learning methods can discover hidden physical patterns in data. By constructing an
end-to-end surrogate model between state parameters and dynamic results, many complex engineering problems such as strong
coupling, nonlinearity, and multiphysics can be efficiently solved. In the field of highly nonlinear explosion and shock
dynamics, a classic detonation driving problem was chosen as the research object. Using numerical simulation results as
training data for machine learning surrogate models, and combining forward simulation and reverse design organically. Based
on deep neural network technology, an end-to-end surrogate model was constructed between feature position velocity profiles,
material dynamic deformation, and engineering factors. And the calculation accuracy of the surrogate model was provided,
verifying the ability to invert engineering factors from velocity profiles. The research results indicate that the end-to-end
surrogate model has high predictive ability, with relative errors of less than 1% in both velocity profile prediction and
engineering factor estimation. It can be applied to the rapid design, high-precision prediction, and agile iteration of highly
nonlinear explosion and impact dynamics problems.
Keywords: computational explosion mechanics; detonation drive; artificial intelligence; machine learning; end-to-end
surrogate model; deep neural network

人工智能（artificial intelligence，AI）是能够和人一样进行感知、认知、决策和执行的人工程序或系
统，是新一轮科技革命和产业变革的重要驱动力量。我国在人工智能方面取得了长足进步，集群作战、

* 收稿日期： 2024-04-10；修回日期： 2024-08-15
第一作者： 柏劲松（1968－ ），男，博士，研究员，bjsong@foxmail.com
通信作者： 刘 洋（1987－ ），男，博士，助理研究员，blonster@163.com

051101-1