Author(s): Joaquin Irazabal; Fernando Salazar; Jose Milla; Javier Garcia; David J. Vicente
Linked Author(s): Fernando Salazar González
Keywords: Large language models (LLMs); CFD; Dam hydraulics; OpenFOAM; AI-based solutions
Abstract: OpenFOAM is one of the most powerful open-source frameworks for Computational Fluid Dynamics (CFD), widely used in research environments. However, its adoption in engineering practice remains limited due to the need for advanced programming skills and a deep understanding of numerical methods and fluid mechanics. To address this gap, we present the EASY-HYDRO project, which main goal is the development of an engineering-oriented software platform using agentic Large Language Models. The resulting interface is capable of guiding engineers through typical modelling tasks–geometry import, mesh generation, boundary condition assignment and parameter definition–reducing the need for direct code manipulation. Several OpenFOAM simulations related to dam hydraulics are presented to illustrate the robustness and flexibility of this innovative tool.
Year: 2026