Residual-guided AI-CFD hybrid method enables stable and scalable simulations: from 2D benchmarks to 3D applications
This provides a scalable and robust hybrid method for engineering applications, addressing a long-standing practical barrier in fluid dynamics simulation.
The paper tackled the instability and lack of automation in hybrid AI-CFD methods for fluid dynamics, achieving stable simulations for over 10,000 timesteps with speedups up to 4.98× and errors as low as ~1E-3 in thermal fields.
Purely data-driven surrogates for fluid dynamics often fail catastrophically from error accumulation, while existing hybrid methods have lacked the automation and robustness for practical use. To solve this, we developed XRePIT, a novel hybrid simulation strategy that synergizes machine learning (ML) acceleration with solver-based correction. We specifically designed our method to be fully automated and physics-aware, ensuring the stability and practical applicability that previous approaches lacked. We demonstrate that this new design overcomes long-standing barriers, achieving the first stable, accelerated rollouts for over 10,000 timesteps. The method also generalizes robustly to unseen boundary conditions and, crucially, scales to 3D flows. Our approach delivers speedups up to 4.98$\times$ while maintaining high physical fidelity, resolving thermal fields with relative errors of ~1E-3 and capturing low magnitude velocity dynamics with errors below 1E-2 ms-1. This work thus establishes a mature and scalable hybrid method, paving the way for its use in real-world engineering.