Physics Informed Neural Networks for PDE Coalescence

Zhang yu

doi:10.47297/taposatWSP2633-456929.20260705

PDF
Export
Share
Collection
Album

Physics Informed Neural Networks for PDE Coalescence

Volume 7, Issue 5

Views: 0 Downloads: 0

Physics Informed Neural Networks for PDE Coalescence
Physics Informed Neural Networks for PDE Coalescence
College of Science, China Jiliang University, Hangzhou, Zhejiang, 310018, China
Vol. 7, Issue 5, Pages: 148-151(2026)
Online First：15 July 2026，

Published：30 May 2026
DOI：10.47297/taposatWSP2633-456929.20260705
Accepted：

Scan QR Code

Zhang yu.Physics Informed Neural Networks for PDE Coalescence[J].Theory and Practice of Science and Technology,2026,07(05):148-151.
DOI：

Zhang yu.Physics Informed Neural Networks for PDE Coalescence[J].Theory and Practice of Science and Technology,2026,07(05):148-151. DOI： 10.47297/taposatWSP2633-456929.20260705.

摘要

Abstract

Physics-Informed Neural Networks (PINNs)

emerging from the convergence of artificial intelligence and scientific computing

are increasingly transforming the solution paradigm for partial differential equations (PDEs). This study examines the fundamental principles

implementation procedures

and application value of PINN-based PDE fusion methods. In response to growing industrial demands associated with multiphysics coupling and complex engineering simulation

the current development status of PDE fusion technologies is systematically reviewed. Several key challenges—including training instability

limited solving efficiency

and insufficient coordination between data-driven learning and physical constraints—are further discussed. On this basis

optimization strategies involving physical constraint enhancement

computational efficiency improvement

and collaborative integration between data and governing physical laws are proposed

providing methodological support for the intelligent solution of complex PDEs.

关键词

Keywords

references

Duan Y ,Tian J., & Wang Y. (2026). Physics-informed neural networks for flow-heat transfer coupling problems: A comprehensive review of methods, applications, and future perspectives [J]. International Journal of Heat and Mass Transfer, (267): 128907-128907.

Lu B ,Mou C., & Lin G. (2026). iPINNER: An iterative physics-informed neural network with ensemble Kalman filter [J]. Journal of Computational Physics, (548): 114592-114592.

Baniya S., & Maity D. (2025). A comprehensive review of theoretical concepts and advancements in physics-informed neural networks with applications in structural engineering [J]. Artificial Intelligence Review, 59 (2): 49-49.

Yang C ,Shan M , et al. (2025). Improved physics-informed neural networks incorporating lattice Boltzmann method optimized by tanh robust weight initialization [J]. Chinese Physics B, 34 (11): 110701-110701.

Zhao Z ,Wang Y., & Zhang W , et al. (2025). Physics-informed neural networks in heat transfer-dominated multiphysics systems: A comprehensive review [J]. Engineering Applications of Artificial Intelligence, (157): 111098-111098.