Port-Hamiltonian Formulation of Coupled Heat Transfer in Cooling Channels of Gas Turbines
My name is Jens Jäschke and I am a third-year PhD-student in the Applied Mathematics and Numerical Analysis (AMNA) group of the University of Wuppertal (BUW). I am also part of the GivEn project, a a cooperation of the Universities of Wuppertal, Trier and Kaiserslautern with Siemens AG and the German Aerospace Center (DLR). In this project, we’re developing a novel multicriteria optimization process that takes the coupling structure of the underlying multiphysical processes into account.
Why I became a PhD student in the AMNA Group
I wasn’t always a mathematician, but started out studying physics for my B.Sc at the TU Darmstadt. Just like approximately half the freshmen in physics, I had big plans and even bigger dreams: Study particle physics, get a PhD, do cool experiments at CERN and find a new elementary particle. However, it turned out that particle physics is complicated and I’m much better at finding problems in the experimental setup than I am at finding particles. I just wasn’t cut out for experimental physics, but I discovered my interest in scientific computer simulations.
So for my master’s degree, I moved to Wuppertal and enrolled in the international course of Computer Simulation in Science, where I came into contact with the AMNA group.
While some of my initial dreams turned out to be unrealistic, my desire to get a PhD and to do some real scientific research never left me. So when I heard about a new project that would involve simulations and allowed my to leverage my background in physics, I immediately applied and was happy to be accepted as a PhD-student in the AMNA group and as a part of the GivEn project.
What I will do during my PhD
While the GivEn project as a whole is concerned with shape optimization of turbine blades, my part of the project is concerned with the mathematical models describing the physical processes in the blade, and especially their coupling structure. Heat flows from the burning hot gas surrounding the blade through the blade into the relatively colder air running through small cooling channels within the blade in highly turbulent flows. Ensuring convergence and stability of the resulting coupled system will be no easy task, but one we hope to achieve through modeling it as a port-Hamiltonian system.
In addition, we will also have to deal with differing dimensions of the subsystems; due to the high turbulence and complex geometries, the cooling channels are generally treated as a one-dimensional flow, since a full simulation would be infeasible.