Dr. -Ing. Björn Stelzner
Dr. Wenkun Zhu

Good written and verbal communication skills in English.

Location: CS

Motivation:

In the field of fluid mechanics, traditional turbulence models such as those based on Reynolds-averaged Navier–Stokes (RANS) equations play a crucial role in solving numerous problems. However, their accuracy in complex scenarios is often limited due to inherent assumptions and approximations, as well as imprecise coefficients in the turbulence model equations. Wall-bounded turbulence plays a critical role in various engineering applications, including engine walls, aircraft wings, and submarine surfaces. However, current hardware facilities are still unable to directly achieve direct numerical analysis in high Reynolds numbers within our acceptable computing time and computing resources. Optical measurement techniques have become essential in turbulence research, enabling non-intrusive acquisition of key physical parameters such as velocity, temperature, and gas concentration. Both high-precision Optical data and RANS simulation seem to become a reasonable choice to analyze the turbulent characteristics near the wall. In our project, Particle Image Velocimetry (PIV) and Planar Laser-Induced Fluorescence (PLIF) will be used to simultaneously measure velocity fields, concentration distributions, and temperature profiles near the wall.

Objectives:

We offer two thesis directions to choose from:

1. Experiment: Interaction between Turbulence and Scalar Transport near the wall using 2D-PIV and Tracer PLIF
The candidate will perform experimental studies using laser systems, high-speed imaging, and advanced data analysis techniques to investigate velocity and scalar transport near the wall. We are currently working on dual scalar measurements of temperature and concentration.

2. Machine Learning: Physics-Informed Neural Networks (PINN) based on Laser Diagnostics
To combine PIV/PLIF experimental data with state-of-the-art PINN frameworks for predicting scalar transport under unknown conditions, we can create a hybrid data-physics model. This model leverages the strengths of both physics-based equations (advection-diffusion PDE) and data-driven learning (from PIV/PLIF).

Responsible:

Prof. Dr.-Ing. Dimosthenis Trimis