A Convex Route to Thermomechanics: Learning Internal Energy and Dissipation

📰 ArXiv cs.AI

Physics-based neural network framework for discovering constitutive models in thermomechanics

advanced Published 31 Mar 2026
Action Steps
  1. Define internal energy and dissipation potential as primary constitutive functions
  2. Express these functions in terms of deformation and entropy
  3. Avoid enforcing mixed convexity-concavity conditions using the proposed framework
  4. Apply the framework to discover constitutive models in fully coupled thermomechanics
Who Needs to Know This

ML researchers and engineers working on physics-informed neural networks can benefit from this framework to model complex thermomechanical systems, and software engineers can apply this knowledge to develop more accurate simulations

Key Insight

💡 Using internal energy and dissipation potential as primary constitutive functions simplifies the modeling of thermomechanical systems

Share This
💡 Physics-informed neural networks for thermomechanics!
Read full paper → ← Back to Reads