arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2309.07899cs.LGcs.NAmath.NAphysics.comp-ph
收藏

通过硬约束改进物理信息驱动的DeepONets

Improving physics-informed DeepONets with hard constraints

University of Hamburg(汉堡大学);Memorial University of Newfoundland(纽芬兰纪念大学);Silesian University in Opava(俄帕瓦斯莱西亚大学);Institute of Mathematics of NAS of Ukraine(乌克兰国家科学院数学研究所)

Rüdiger Brecht, Dmytro R. Popovych, Alex Bihlo, Roman O. Popovych

AI总结:

针对现有物理信息驱动神经网络需学习微分方程初边值条件的问题,提出带硬约束的改进策略,使初边值条件在解中精确表示,且保证深度算子网络多步时间步进求解初值问题时所得函数至少连续。

摘要:

当前的物理信息驱动(标准或深度算子)神经网络仍依赖于准确学习其求解的微分方程组的初始和/或边界条件。相比之下,标准数值方法在计算中纳入这类条件时无需对其进行学习。本研究提出改进现有物理信息驱动的深度学习策略,使初始和/或边界条件无需被学习,且能在预测解中被精确表示。此外,该方法保证了当深度算子网络被多次应用于对初值问题的解进行时间步进时,所得函数至少是连续的。

英文摘要:

Current physics-informed (standard or deep operator) neural networks still rely on accurately learning the initial and/or boundary conditions of the system of differential equations they are solving. In contrast, standard numerical methods involve such conditions in computations without needing to learn them. In this study, we propose to improve current physics-informed deep learning strategies such that initial and/or boundary conditions do not need to be learned and are represented exactly in the predicted solution. Moreover, this method guarantees that when a deep operator network is applied multiple times to time-step a solution of an initial value problem, the resulting function is at least continuous.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑