arXiv:2606.28119v1 Announce Type: cross Abstract: We introduce a physics-constrained neural network (PCNN) for the rapid prediction of rigorous coupled-wave analysis (RCWA) outputs in the form of Jones matrices. Starting from energy conservation in lossless layered periodic structures, we use the fact that RCWA outputs lie on a Stiefel manifold. This energy constraint is enforced as a hard condition by projecting onto the manifold using differentiable symmetric orthogonalization. The resulting surrogate enforces energy conservation by construction while preserving differentiability for gradien
The increasing complexity of scientific modeling, particularly in physics and engineering, necessitates faster and more accurate simulation methods, driving innovation in AI-driven surrogates.
This development offers a method to create highly accurate and physics-consistent AI models for complex systems, potentially accelerating R&D in physical sciences and engineering.
Traditional computational physics models can now be significantly augmented or replaced by AI surrogates that inherently respect physical laws, leading to more efficient design and analysis cycles.
- · AI/ML researchers
- · Optics and photonics industries
- · Material science
- · Semiconductor design
- · Traditional RCWA software companies without AI integration
- · Researchers relying solely on purely data-driven, non-physics-constrained AI mod
Faster and more accurate simulation for optical and periodic structures will reduce design costs and time.
This methodology could be generalized to other physics domains, leading to a broader adoption of physics-constrained AI for scientific discovery and engineering.
The acceleration of material and device discovery enabled by such AI tools could lead to breakthroughs in energy efficiency, quantum computing, or advanced sensors.
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