Discovering Thermodynamically Admissible Dissipation Potentials via Grammar-Based Symbolic Regression
arXiv:2605.31532v1 Announce Type: cross Abstract: Constitutive laws for inelastic materials must satisfy strict thermodynamic admissibility requirements, yet current data-driven approaches sacrifice interpretability, even when formal guarantees are provided by physics-encoded architectures. We propose a symbolic regression framework for the data-driven discovery of dissipation potentials governing the evolution of internal variables within the Generalized Standard Materials (GSM) formalism. Starting from the Clausius--Duhem inequality, we enforce the thermodynamic requirements, convexity and n
The increasing maturity of AI and symbolic regression techniques is enabling new approaches to complex scientific and engineering problems previously constrained by traditional modeling methods.
This development represents a step towards more interpretable and thermodynamically sound AI for material science, which is critical for safety-critical applications and advanced manufacturing.
The ability to discover interpretable constitutive laws for inelastic materials using data-driven methods, while adhering to physical constraints, could accelerate material design and understanding.
- · Material scientists
- · Engineers in advanced manufacturing
- · AI researchers in scientific discovery
- · Traditional empirical material modeling approaches
Improved simulation accuracy and design of complex materials.
Faster development cycles for new alloys, composites, and other engineering materials.
Potential for AI-driven autonomous material discovery and optimization, reducing reliance on extensive physical experimentation.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG