Accelerating multicomponent phase-coexistence calculations with physics-informed neural networks
Satyen Dhamankar, Shengli Jiang, Michael A. Webb
Abstract
We develop a physics-informed machine learning workflow that accelerates multicomponent phase-coexistence calculations on the number, composition, and abundance of phases. The workflow is demonstrated for systems described by Flory–Huggins theory.
Topics & Concepts
Artificial neural networkPhase (matter)Statistical physicsPhysicsComputer scienceTheoretical physicsArtificial intelligenceQuantum mechanicsQuantum, superfluid, helium dynamicsModel Reduction and Neural NetworksPower Transformer Diagnostics and Insulation