German Success Stories in Industrial Mathematics

Frontmatter

From Probabilistic Prediction of Fatigue Life to a New Design Approach for Gas Turbines

Abstract

Gas turbines are used in aviation and energy production. As efficiency of gas turbines increases with firing temperatures, the hot gas components of a gas turbine are subject to extreme termo-mechanical load.

Tilmann Beck, Hanno Gottschalk, Rolf Krause, Georg Rollmann, Sebastian Schmitz, Lucas Mäde

Increasing the Reliability of Multi-megawatt Gas Compressors

Abstract

According to our industrial partner “interruption in the gas supply can mean financial losses that range into the hundreds of millions USD per year for large facilities”. Fast methods and soft- ware for Nonlinear Model Predictive Control (NMPC) on embedded systems greatly improve the reliability of multi-megawatt gas compressors in comparison with less advanced control strategies in the face of power disturbances.

Hans Georg Bock, Christian Kirches, Andreas Potschka, Thomas Besselmann, Sture Van de moortel

Mathematical Modeling, Simulation and Optimization for CERN‘s Quench Protection System

Abstract

Superconducting magnets are used to generate high magnetic fields and are employed in several applications, such as in particle accelerators to control the beam of particles that is travelling through them. The superconducting material can, under certain circumstances, quench, that is, lose its superconductivity and as a consequence get potentially highly damaged.

Idoia Cortes Garcia, Sebastian Schöps, Lorenzo Bortot, Matthias Mentink

Efficient Gas Network Simulations

Abstract

Simulations of the gas network infrastructure play an important role in energy supply and the green energy transition. Especially volatilities induced by renewable energies increase the need for more transient simulations in shorter time-spans.

Christian Himpe, Sara Grundel, Peter Benner

Capacity Evaluation for Large-Scale Gas Networks

Abstract

Natural gas is important for the energy turnaround in many countries like in Germany, where it serves as a bridging energy towards a fossil-free energy supply in the future. About 20% of the total German energy demand is provided by natural gas, which is transported through a complex pipeline network with a total length of about 30000 km, and the efficient use of the given transport infrastructure for natural gas is of political, economic, and societal importance.

Martin Schmidt, Benjamin Hiller, Thorsten Koch, Marc E. Pfetsch, Björn Geissler, René Henrion, Imke Joormann, Alexander Martin, Antonio Morsi, Werner Römisch, Lars Schewe, Rüdiger Schultz, Marc C. Steinbach

Purity Assessment of Pellets using Deep Learning

Abstract

Plastic pellets—an unknown, but key ingredient for virtually anything made of plastic such as extra-high voltage cables, medical technology or advanced material. This modern technology, however, requires ever increasing purity of plastic granulate.

Jens Behrmann, Maximilian Schmidt, Jannik Wildner, Peter Maass

Modelling and Simulation of High-Frequency Induction Welding

Abstract

Almost all manufacturing sectors, from construction through transport to consumer goods, are largely based on the utilisation of steels.Steel products often compete favourably with alternative material solutions in cost efficiency, life cycle analyses

Prerana Das, Dietmar Hömberg

Digital Material Characterization

Abstract

Fiber reinforced plastics have a high stiffness to weight ratio and can be produced in a cost-effective way on a mass production scale by injection or compression molding. Therefore, this type of material plays an important role for producing lightweight components.

Matthias Kabel, Jonathan Köbler, Heiko Andrä

Virtual Product Development and Digital Validation in Automotive Industry

Abstract

Engineers working with software tools for computer aided design (CAD), digital mock-up and virtual assembly need to handle rigid as well as flexible parts at interactive rates. Slender flexible parts like cables and hoses are important for the functionality of complex mechatronic machinery like cars, trucks or aircraft.

Joachim Linn, Fabio Schneider, Klaus Dreßler, Oliver Hermanns

Nonwoven Production Processes

Abstract

The application area of nonwoven production, embedded in the field of fluid-structure interactions, offers a multitude of mathematical challenges, since the complexity of the processes renders them unamenable to standard simulation techniques.

Nicole Marheineke, Raimund Wegener

Optimization of Multimaterial Dies via Numerical Simulations

Abstract

Extrusion processes and the corresponding optimization of the extrusion die belongs to the great challenges in polymer processing and is of great importance for the resulting multimaterial extrudates as industrial products, regarding economical aspects as well as their environmental impact.

Otto Mierka, Markus Geveler, Stefan Turek

Energy Conservation for Wine Fermentation

Abstract

A large amount of energy is used in the production of wine, which has to be cooled during fermentation in order to produce a multiplicity of aromas. If not cooled, the added yeast transforms the sugar content of the raw material, the must, into alcohol without much of a taste, in very short time. Thus the cooling during the fermentation process is a major component in the production process of wine.

Jan Bartsch, Alfio Borzi, Jonas Müller, Christina Schenk, Dominik Schmidt, Volker Schulz, Kai Velten

Modelling and Simulation of Tank Mixing

Abstract

The beverage and wine industries routinely conduct blending of various liquids or homogenization of additives in large vessels equipped with fast-rotating propeller mixers. Mixing times, i.e. propeller run times, are usually based on empirical experiences. Mathematical modeling and simulation could be used to predict mixing times in such situations, but common methods either provide only approximate solutions or are computational unfeasible.

Jonas Müller, Dominik Schmidt, Kai Velten

Personalized Medicine—Optimized EPO Dosing

Abstract

Currently, epoetin alfa (EPO) doses for hemodialysis patients are usually prescribed using dosing protocols, where doses can be looked up based on previous hemoglobin (Hgb) measurements. The aim is to stabilize Hgb levels within the narrow target window of 10 – 12 g/dl. This is a delicate objective, as patients’ response to treatment varies highly and doses should be as small as possible to mitigate drug-related hazards.

Franz Kappel, Stefan Volkwein

Wear Testing of Knee Implants

Abstract

Hardware wear tests for knee implants, necessary for market authorization, are expensive and time-consuming. Faster than realtime wear simulations have the potential to speed up the design process for implants by providing frequent and early feedback.

Ralf Kornhuber, Oliver Sander, Anton Schiela, Martin Weiser

Radiotherapy Treatment Planning

Abstract

Radiotherapy treatment is one of the most important cancer therapies, as approximately half of all cancer patients are treated with radiation in some form during their treatment. Planning radiation treatment is challenging because the potential success of the therapy has to be weighed against potential unwanted side effects, and treatment is highly individual: it is a priori unknown what the chances and limitations of the therapy are for a particular patient.

Karl-Heinz Küfer, Philipp Süss

MALDI Imaging: Exploring the molecular landscape

Abstract

The driving force behind recent developments in Life Sciences such as drug discovery, individual therapy planning or pathway detection in systems biology are the Omics-technologies (Proteomics, Lipidomics, Metabolomics, etc.). Over the last 15 years these technologies have been partially revolutionized due to the advance of a new bioanalytic methodology called MALDI imaging (matrix assisted laser desorption ionization).

Peter Maass, Lena Hauberg-Lotte, Tobias Boskamp

Energy-Efficient Timetabling in a German Underground System

Abstract

Timetabling of railway traffic and other modes of transport is among the most prominent applications of discrete optimization in practice. However, it has only been recently that the connection between timetabling and energy consumption has been studied more extensively.

Andreas Bärmann, Patrick Gemander, Alexander Martin, Maximilian Merkert

Mathematical Optimization and Machine Learning for Efficient Urban Traffic

Abstract

Traffic jams cause economical damage, which has been estimated between 10 and 100 billion Euros per year in Germany.

Johanna Bethge, Rolf Findeisen, Do Duc Le, Maximilian Merkert, Sebastian Sager, Anton Savchenko, Hannes Rewald

Train Rotation Optimization

Abstract

Train rotation planning, i.e., the construction of efficient circulations of rolling stock in order to implement a timetable, is a key problem for railway companies. A major challenge is the formation of trains from several individual units of rolling stock, which can differ by type, position, and orientation.

Ralf Borndörfer, Markus Reuther, Steffen Weider, Peter Schütz, Kerstin Waas

MESHFREE Simulations in Car Design: Closing the Gaps of Classical Simulation Tools

Abstract

Numerical simulation speeds up the design process of products tremendously. In numerous cases, simulations cannot cover the complete spectrum of the product’s functionalities and its production processes. Usually, there are modeling tasks, for which classical simulation tools are unfeasible as they require too much human and computation resources, or fail completely.

Jörg Kuhnert

PASIROM: Parallel Simulation and Robust Optimization of Electro-Mechanical Energy Converters

Abstract

Digital twins are used in industry to understand the life cycle of engineering products. Particularly during product design, simulations are important, e.g., to optimize the geometry or to investigate possible sources of uncertainties. The mathematical model behind the twin is typically a set of partial differential equations.

Iryna Kulchytska-Ruchka, Sebastian Schöps, Michael Hinze, Stephanie Friedhoff, Stefan Ulbrich

Efficient Virtual Design and Testing of Autonomous Vehicles

Abstract

With the ever increasing capabilities of sensors and controllers, autonomous driving is quickly becoming a reality. This disruptive change in the automotive industry poses major challenges for manufacturers as well as suppliers as entirely new design and testing strategies have to be developed to remain competitive. Most importantly, the complexity of autonomously driving vehicles in a complex, uncertain, and safety-critical environment requires new testing procedures to cover the almost infinite range of potential scenarios.

Sebastian Peitz, Michael Dellnitz, Sebastian Bannenberg

Tetgen: Tetrahedral Mesh Generation for Complex Simulations

Abstract

Electromagnetic fields in magnetic resonance imaging, groundwater flows, charge carriers in semiconductors and numerous further spatially distributed physical processes have one thing in common: they can be modeled in a sophisticated and efficient way by means of partial differential equations.

Jürgen Fuhrmann, Hang Si

WORHP: Development and Applications of the ESA NLP SOLVER

Abstract

WORHP is a mathematical software library for solving continuous large-scale nonlinear optimization problems numerically.

Matthias Knauer, Christof Büskens

Non-convex Pareto Set navigation

Abstract

Many real world applications require the solution of multi-objective optimization problems. Exploration of the corresponding set of best compromises, i.e., the Pareto set, supports the decision making process.

Dimitri Nowak, Alexander Scherrer, Michael Bortz, Karl-Heinz Küfer

Correction to: Capacity Evaluation for Large-Scale Gas Networks

Martin Schmidt, Benjamin Hiller, Thorsten Koch, Marc E. Pfetsch, Björn Geißler, René Henrion, Imke Joormann, Alexander Martin, Antonio Morsi, Werner Römisch, Lars Schewe, Rüdiger Schultz, Marc C. Steinbach