A Surrogate Deep-Learning Super-Resolution Framework for Accelerating Finite Element Method-Based Fluid Simulations

Key Points

• The research aims to create a deep learning framework that accelerates fluid simulations based on the finite element method.
• Developed a surrogate super-resolution framework using deep learning techniques
• Applied the framework to finite element method-based computational fluid dynamics
• Conducted evaluations to assess computational efficiency and simulation accuracy
• Demonstrated significant acceleration in fluid simulations
• Achieved high-resolution results with reduced computational cost
• Showed improved accuracy in simulations compared to traditional methods