Prof. Dr. Nicolas Gauger, Chair for Scientific Computing (SciComp), TU Kaiserslautern

SciComp Seminar Series

Please contact Prof. Gauger, if you want to register for an online talk in our SciComp Seminar Series or just to register for the seminar.

A list of the already scheduled talks can be found –> here:

Prof. Dr. Bernd Ulmann, FOM University of Applied Sciences in Frankfurt/Main

Title: Analog and Hybrid Computing in the 21st Century

Abstract:

-> Download Slides  
Computation by electronic analogues is gaining more and more interest in many areas of research as well as industrial applications. Being one central pillar of unconventional computing, this particular paradigm offers unmatched advantages with respect to energy efficiency as well as computational power for certain problem domains not limited to the treatment of systems of ODEs and PDEs but also areas such as AI. This talk gives an introduction into modern analog and hybrid computing with examples and an outlook to future application areas.

More about the speaker:

Bernd Ulmann studied mathematics with a minor in philosophy at Johannes Gutenberg University in Mainz. After several years in industry (process automation in oil refineries) he wrote his PhD at Hamburg University about analog computers, his life long favourite topic. He is professor at the FOM university of applied sciences in Frankfurt/Main and is a co-founder of anabrid GmbH where modern, highly integrated analog computers are developed. In his free time he collects, restores and builds analog computers (what else 🙂 ).

How to join online

You can join online via Zoom, using the following link:
https://uni-kl-de.zoom-x.de/j/69269239534?pwd=Z9UOzMpkhMjrxVhll3d49sNHFe9Fd1.1

Dr. Bora Kalpakli, LIKUA, Heidelberg, Germany

Title: NeuralFlow: Learning CFD Solution Operators Through Solver Backpropagation

Abstract:

Computational Fluid Dynamics (CFD) traditionally obtains flow solutions by repeatedly
applying numerical operations that reduce the imbalance in the discretized flow field until
a converged state is reached. Although modern CFD algorithms are accurate and robust,
their computational cost remains a limitation for applications requiring repeated simulations,
large parameter studies, optimization, or near-real-time prediction. NeuralFlow investigates
a different approach: a neural operator is trained not to reproduce a database of previously
computed CFD solutions, but to learn how the CFD solver itself drives a flow field toward
convergence.
NeuralFlow is based on a Graph Neural Operator (GNO) defined directly on an unstructured
finite-volume mesh. Mesh cells are represented as graph nodes and their physical
connections as edges. Starting from an initial flow field, the GNO predicts corrections to the
primitive variables and is applied recurrently over multiple rollout steps. This local graph
formulation allows the same learned operator to act on different parts of an unstructured
mesh while progressively transforming the current state toward a converged flow solution.
A central feature of NeuralFlow is that no reference CFD solution data are required for
training. It is also fundamentally different from a Physics-Informed Neural Network (PINN):
the governing differential equations are not imposed directly on the neural network as a
collocation-based loss. Instead, the predicted flow state is evaluated by the existing CMPS
CFD solver. CMPS performs its normal finite-volume operations, including numerical flux
evaluation, boundary-condition treatment, residual assembly, and Jacobian evaluation. Exact
solver sensitivities are obtained by automatic differentiation and are used to construct the
gradient with respect to the GNO output. This solver-derived gradient is then backpropagated
through the neural network, allowing the GNO weights to learn directly from the behaviour
of the numerical CFD algorithm.
The seminar will present the NeuralFlow architecture, the graph representation of finitevolume
meshes, recurrent rollout training, and the coupling between forward-mode automatic
differentiation inside the CFD solver and reverse-mode backpropagation inside the neural
network. Initial results show that the learned operator can progressively reduce CFD
residuals and reproduce laminar-flow behaviour without supervised solution data. The longterm
objective is a generalizable learned CFD solver that retains the numerical and physical
structure of conventional CFD while transferring part of the iterative solution process to a
trainable graph neural operator.

How to join online

You can join online via Zoom, using the following link:
https://uni-kl-de.zoom-x.de/j/69269239534?pwd=Z9UOzMpkhMjrxVhll3d49sNHFe9Fd1.1