Rui Fang
The Ohio State University
Title
Fast and Accurate Methods for Predicting Fluid Flows
Abstract
Fluids in motion exhibit complex behaviors that are difficult to predict yet critical to applications such as renewable energy and weather forecasting. Numerical simulations can capture these behaviors with high accuracy across a wide range of flow regimes. The Navier–Stokes equations form the fundamental mathematical model for nearly all fluid flows.
Due to the chaotic nature of fluid dynamics, tiny errors in initial conditions can amplify rapidly, leading to finite predictability horizons. To address uncertainties in problem data, we develop ensemble simulations combined with a penalty-based incompressibility relaxation, allowing efficient computation across multiple flow realizations.
Direct Numerical Simulation is often impractical due to its demand for fine spatial and temporal resolution. URANS models offer a practical alternative. The second project investigates a 1/2-equation turbulence model that simplifies the standard 1-equation URANS formulation while maintaining predictive capability.
Data assimilation integrates the governing dynamics with observational data to improve predictability. One remaining challenge is that nudging-based assimilation requires manual tuning of a key parameter. The third project develops a self-adaptive parameter selection method to automate this process and enhance the reliability of forecasts.