首页|An Integral Method for Solving Dynamic Equations with Fluid-Solid Coupling

An Integral Method for Solving Dynamic Equations with Fluid-Solid Coupling

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
In this work,a new methodology is presented to mainly solve the fluid-solid interaction(FSI)equation.This methodology combines the advantages of the Newmark precise integral method(NPIM)and the dual neural network(DNN)method.The NPIM is employed to modify the exponential matrix and loading vector based on the DNN integral method.This involves incorporating the basic assumption of the Newmark-β method into the dynamic equation and eliminating the acceleration term from the dynamic equilibrium equation.As a result,the equation is reduced to a first-order linear equation system.Subsequently,the PIM is applied to integrate the system step by step within the NPIM.The DNN method is adopted to solve the inhomogeneous term through fitting the integrand and the original function with a pair of neural networks,and the integral term is solved using the Newton-Leibniz formula.Numerical examples demonstrate that the proposed methodology significantly improves computing efficiency and provides sufficient precision compared to the DNN method.This is particularly evident when analyzing large-scale structures under blast loading conditions.

Newmark precise integral method(NPIM)Dual neural networks(DNN)Fluid-solid interaction(FSI)Dynamic time-history analysisBlast loading

Xin Zhang、Jie Liu、Pu Xue、Shuowen Yan、Yahao Xu、M.S.Zahran

展开 >

School of Aeronautics,Northwestern Polytechnical University,Xi'an 710072,China

School of Mechatronical Engineering,Beijing Institute of Technology,Beijing 100081,China

Department of Civil Engineering,Military Technical College,Cairo 11435,Egypt

National Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaNational Natural Science Foundation of China

12072288U224127412272319

2024

10.1007/s10338-023-00434-8

固体力学学报(英文版)
中国力学学会

固体力学学报(英文版)

EI
影响因子：0.214
ISSN：0894-9166
年,卷(期)：2024.37(1)
  • 26