PINNs with Differentiable Chemistry Solve Stiff Reaction Systems

Researchers have developed a novel framework integrating a differentiable chemistry solver with physics-informed neural networks (PINNs) to tackle stiff and parameterized reaction systems. This approach addresses limitations of standard PINNs by incorporating a specialized solver, a network architecture for parameterized solutions, and residual weighting optimized for stiff reactions. The framework's effectiveness was demonstrated on hydrogen combustion models, successfully handling initial/boundary value problems, inverse parameter identification, and parameterized partial differential equations, thereby extending PINNs to previously inaccessible chemical systems. AI

影响 Extends the applicability of physics-informed neural networks to complex, stiff chemical systems, potentially enabling new scientific simulations and discoveries.

排序理由 This is a research paper detailing a new framework for solving stiff chemical reaction systems using physics-informed neural networks. [lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

• arXiv
• neural ODEs

• 论文
• 基础设施

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →