Mechanism-Separated Closed-Form Transition Modeling via Field Inversion and Symbolic Regression

📝 Status: Under review (AIAA Journal). The preprint is available on arXiv.

🔍 Overview

This work proposes a laminar-to-turbulent transition model that covers natural, crossflow, and separation-induced transition, with a separate correction branch for each mechanism.

• Data-driven but closed-form: The corrections are built with field inversion (FI) and symbolic regression (SR), giving explicit algebraic corrections to the Spalart–Allmaras production term.
• Low cost at runtime: The model needs no additional transport equations and no neural-network inference.
• Solver-independent: The model was implemented in an independent flow solver and tested on canonical transition cases, a natural-laminar-flow transport wing, and a hovering rotor.

✅ Key Results

• Captures the main transition-front trends and the resulting changes in aerodynamic performance.
• For the hovering rotor, the model needed about 55% of the wall-clock time of a comparable transport-equation transition model.

📄 Preprint

Seunghyun Joo, Younghyo Kim, Kwanjung Yee, “Mechanism-Separated Closed-Form Transition Modeling via Field Inversion and Symbolic Regression,” arXiv:2609.24162 [physics.flu-dyn], 2026. https://arxiv.org/abs/2609.24162

Recommended citation: Seunghyun Joo, Younghyo Kim, Kwanjung Yee. (2026). “Mechanism-Separated Closed-Form Transition Modeling via Field Inversion and Symbolic Regression.” Under review, AIAA Journal. arXiv:2609.24162.
Download Paper