paper-with-me

홈 › Papers

Neural PDE Solvers with Physics Constraints: A Comparative Study of PINNs, DRM, and WANs

2025-10-09 · Jiakang Chen arxiv

Partial differential equations (PDEs) underpin models across science and engineering, yet analytical solutions are atypical and classical mesh-based solvers can be costly in high dimensions. This dissertation presents a unified comparison of three mesh-free neural PDE solvers, physics-informed neural networks (PINNs), the deep Ritz method (DRM), and weak adversarial networks (WANs), on Poisson problems (up to 5D) and the time-independent Schrödinger equation in 1D/2D (infinite well and harmonic oscillator), and extends the study to a laser-driven case of Schrödinger's equation via the Kramers-Henneberger (KH) transformation. Under a common protocol, all methods achieve low $L_2$ errors ($10^{-6}$-$10^{-9}$) when paired with forced boundary conditions (FBCs), forced nodes (FNs), and orthogonality regularization (OG). Across tasks, PINNs are the most reliable for accuracy and recovery of excited spectra; DRM offers the best accuracy-runtime trade-off on stationary problems; WAN is more sensitive but competitive when weak-form constraints and FN/OG are used effectively. Sensitivity analyses show that FBC removes boundary-loss tuning, network width matters more than depth for single-network solvers, and most gains occur within 5000-10,000 epochs. The same toolkit solves the KH case, indicating transfer beyond canonical benchmarks. We provide practical guidelines for method selection and outline the following extensions: time-dependent formulations for DRM and WAN, adaptive residual-driven sampling, parallel multi-state training, and neural domain decomposition. These results support physics-guided neural solvers as credible, scalable tools for solving complex PDEs.

📄 PDF Abstract BibTeX arXiv:2510.09693

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Soliton profiles: Classical Numerical Schemes vs. Neural Network - Based Solvers

2025-12-31 · Chandler Haight, Svetlana Roudenko, Zhongming Wang arxiv

We present a comparative study of classical numerical solvers, such as Petviashvili's method or finite difference with Newton iterations, and neural network-based methods for computing ground states or profiles of solita…

Computational Efficiency

Limitations of Physics-Informed Neural Networks: a Study on Smart Grid Surrogation

2025-08-29 · Julen Cestero, Carmine Delle Femine, Kenji S. Muro, Marco Quartulli 외 arxiv

Physics-Informed Neural Networks (PINNs) present a transformative approach for smart grid modeling by integrating physical laws directly into learning frameworks, addressing critical challenges of data scarcity and physi…

Trajectory Prediction

BO-SA-PINNs: Self-adaptive physics-informed neural networks based on Bayesian optimization for automatically designing PDE solvers

2025-04-14 · Rui Zhang, Liang Li, Stéphane Lanteri, Hao Kang 외

Physics-informed neural networks (PINNs) is becoming a popular alternative method for solving partial differential equations (PDEs). However, they require dedicated manual modifications to the hyperparameters of the netw…

Bayesian Optimization

NewPINNs: Physics-Informing Neural Networks Using Conventional Solvers for Partial Differential Equations

2026-01-23 · Satish Chandran, Maedeh Makki, Maziar Raissi, Adrien Grenier 외 arxiv

We introduce NewPINNs, a physics-informing learning framework that couples neural networks with conventional numerical solvers for solving differential equations. Rather than enforcing governing equations and boundary co…

Hypernetwork-based Meta-Learning for Low-Rank Physics-Informed Neural Networks

2023-10-14 · NeurIPS 2023 11

In various engineering and applied science applications, repetitive numerical simulations of partial differential equations (PDEs) for varying input parameters are often required (e.g., aircraft shape optimization over m…

Meta-Learning