Kolloquium Wintersemester 2015/16

Dr. Mikhail Lapine, University of Technology, Sydney
Einladender: Dr. J. Imbrock

The extravagant field of metamaterials has been bubbling with bright ideas and promising designs for more than a decade of its dramatic progress. I will present a brief conceptual overview of the origins and development of nonlinear metamaterials, addressing the basics of theoretical approaches, key ideas on implementation, and certain interesting phenomena. Particular emphasis will be given to the most recent development, including the use of new degrees of freedom in metamaterial design, and the arising multi-physics of their response. Most recent examples include various aspects of electromagneto—mechanical nonlinearities, artificial electrostriction, nonlinear 'traffic lights' and phase matching with hyperbolic metamaterials.

Unfitted finite element methods using bulk meshes for surface partial differential equations

Dr. Thomas Ranner, University of Leeds
Einladender: Prof. Dr. C. Engwer

I introduce a family of novel finite element methods for partial differential equations on surfaces. The key idea is that the finite element space is based on continuous piecewise linear finite element functions on a bulk triangulation which is independent of the surface. I will present robust numerical analysis for a simple model elliptic problem and provide computational examples to show the flexibility and efficiency of the methods to the evolving and coupled bulk-surface cases.

Unbalanced Optimal Transport: Dynamic and Static Perspective

Dr. Bernhard Schmitzer, CEREMADE, Université Paris-Dauphine
Einladender: Prof. Dr. B. Wirth

Today Optimal transport is a popular modelling tool in image analysis and other fields of applied mathematics. Its original formulation dates back to Monge. There is a `static' linear programming formulation due to Kantorovich and more recently a `time-dynamic' perspective has been introduced by Benamou & Brenier. While the former is usually the foundation for efficient algorithms the latter is a popular basis for constructing more general models that also allow for creation and annihilation of mass.

We introduce a class of generalizations of the original dynamic problem and prove equivalence with a corresponding generalized class of static problems. This could allow for combining the advantages of efficient algorithms with more general models. A particular candidate in this framework will be discussed: the interpolation between the Fisher-Rao and the Optimal Transport metric, which we believe to be well-suited for the modelling of natural growth processes.

Based on joint work with Lenaic Chizat, Gabriel Peyré and François-Xavier Vialard

In most of the problems of interest involving complex systems (physico-chemical, biological, financial, sociological, etc.) the fluctuations of the quantities characterizing those systems are essential. Thus, a probabilistic description of their dynamics is needed.
In this course, we will discuss first how noise arises even in a simple Hamiltonian system when one concentrates on a subset of relevant variables, rather than on the full phase space. Dealing with noise requires the use of probabilistic ideas. We will introduce the concept of stochastic process in general and Markov processes in particular. We present three descriptions of Markov processes: the Master equation, the Langevin equation and the Fokker-Planck equation. After discussing the theoretical ideas behind those descriptions we apply them to some simple cases amenable to analytical solutions. Unfortunately, most of the interesting realistic problems can not be tackled with pure analytical tools. Numerical simulation methods become essential to deal with realistic problems. We discuss some of the numerical simulation procedures.

Monday, 2.11.15, 16:00-18:00
Wednesday, 4.11.15, 16:00-18:00
Thursday, 4.11.15, 16:00-18:00
Friday, 6.11.15, 14:00-16:00

The lecture takes place in room HS 404 of the institute for theoretical physics, Wilhelm-Klemm-Str. 9

Dr. Ingo Holtkötter, Schimmelpfennig + Becke GbR, Ingenieurbüro für Unfallrekonstruktion, Münster
Einladender: Dr. O. Kamps

Unfälle im Straßenverkehr sind an der Tagesordnung, obwohl es für die meisten Beteiligten ein einzigartiges Erlebnis in ihrem Leben ist. Eine objektive Beurteilung des Unfallgeschehens aus der Erinnerung heraus ist nur schwer möglich, da der kurze Zeitraum von oftmals zwei bis drei Sekunden, in denen sich ein Unfall entwickelt, mit ungewöhnlichen Eindrücken überladen ist.

Für den Unfallanalytiker hingegen stellt dieser kurze Zeitraum den Kern seiner Arbeit dar. Im Fachgebiet der Unfallrekonstruktion geht es darum, den Unfallhergang in seinen wesentlichen Details zu rekonstruieren und für den Nichttechniker in nachvollziehbarer Weise darzustellen.

Hierzu werden sowohl die Spurendokumentation an der Unfallstelle und die technische Untersuchung der Unfallfahrzeuge als auch Crashversuche und Computersimulationen erangezogen, um objektive und gerichtsfeste Gutachten zu erstellen, die die Basis für Regulierungsentscheidungen und Gerichtsurteile bilden.

Der Vortrag bietet eine Übersicht über die Methoden der Unfallrekonstruktion, stellt das Berufsbild des Unfallanalytikers als Schnittstelle zwischen Polizei, Versicherungen, Beteiligten und Gerichten vor und zeigt auf, wie trotz Computersimulationen auch heute noch in Münster täglich Crashversuche durchgeführt werden, unfallanalytische Gutachten auf eine solide Basis zu stellen. Dabei werden nicht nur Straßenverkehrsunfälle, sondern auch Spiel-/Sport- und Arbeitsunfälle behandelt, die häufig sehr spezielle Untersuchungen und Versuche erfordern.

Dr. rer. nat. Ingo Holtkötter ist öffentlich bestellter und vereidigter Sachverständiger für Straßenverkehrsunfälle sowie Sachverständiger für Unfälle mit mechanisch-technischem Gerät im Ingenieurbüro Schimmelpfennig und Becke (Münster). Nach dem Physikstudium und der Promotion am physikalischen Institut in Münster ist er dem Themengebiet Kollisionsmechanik treu geblieben. Nur, dass nichtmehr spinpolarisierte Elektronen auf Moleküle geschossen und mit Quantenmechanik beschrieben werden, sondern Fahrzeuge gecrasht werden und klassische Mechanik zum Einsatz kommt …

- suppression of the mosquito population,
- reduction of the disease transmissibility.

This presentation will be focused on implementation of these external control actions using the frameworks of mathematical modeling and control theory approach. In the first part, I will resent and endemo-epidemic model derived from registered dengue case in Cali, Colombia and then propose a set of optimal strategies for dengue prevention and control. In the second part, I will present an alternative and unconventional vector control technique based on the use of biological control agent (Wolbachia) and formulate a decision-making model for Wolbachia ransinfection in wild Aedes aegypti populations.

References:

• [1] Brunner, D., Soriano, M. C., Mirasso, C. R., & Fischer, I. , Parallel photonic information  processing at gigabyte per second data rates using transient states. Nature Communications, 4, 1364 (2013).
• [2] Brunner, D., Soriano, M. C., & Fischer, I. , High-Speed Optical Vector and Matrix Operations Using a Semiconductor Laser. IEEE Photonics Technology Letters, 25(17), 1680–1683 (2013).
• [3] Appeltant, L., Soriano, M. C., Van der Sande, G., Danckaert, J., Massar, S., Dambre, J., … Fischer, I. , Information processing using a single dynamical node as complex system. Nature Communications, 2, 468 (2011).
• [4] Soriano, M.C., Brunner, D., Escalona-Moran, M., Mirasso, C.R. and Fischer, I., Minimal approach to neuro-inspired information processing. Front. Comput. Neurosci. 9:68. doi: 10.3389/fncom.2015.00068 (2015).

Data assimilation for a stochastic urban crime model and equation-free continuation

Dr. David Lloyd, University of Surrey, Department of Mathematics
Einladender: Prof.Dr. U. Thiele

In this talk we introduce several of the novel mathematical challenges one has to face when trying to analyse, model, incorporate and forecast crime data. In order to elucidate the issues involved, we concentrate on burglary crime and the LA model that has links with the PREDPOL model used by the LA Police Department. Using a combination of stochastic modelling, dynamical systems analysis, stochastic bifurcation analysis/equation-free techniques and novel filtering methods, we show how one can in principle carry out data assimilation for this set-up. In doing, so we start to raise more fundamental sociological and societal issues that requires a Steering Complex Adaptive Systems Methodology. This is work done with Naratip Santitissadeekorn (Surrey) and Martin B. Short (Georgia Tech.)

Dr. Andrey Iljin, Institute of Physics, National Academy of Sciences of Ukraine, Kiev, Ukraine
Einladende: Prof. Denz

Liquid crystals (LC) acquired their popularity due to large birefringence and easy response to applied electric or light field. The LC director reorientation results in substantial changes of the effective refractive index of the LC medium providing efficient control of the optical performance of the LC cell, with one of its lonely drawbacks being a rather slow relaxation.
Apart from quite comprehensive and ingenious tricks aimed to bring the response times of orientational processes down, an alternative approach has been gaining particular attention recently. Influencing the order parameter of LC one can realise switching with the characteristic times being several orders of magnitude shorter than that of the LC director reorientation with good prospects in nonlinear photonics.

Nonlinear dynamics of systems with time-varying delay and its relevance for modern machining processes

Prof. Günter Radons, TU Chemnitz
Einladender: Dr. O. Kamps


Delay differential equations provide well-known models for many processes in nature and technology. While models with constant delay are a reasonable and much investigated starting point for many applications, it is also clear that in reality delays fluctuate naturally and often can be influenced to vary systematically. Despite its high, practical relevance, the consequences of such time-varying delays is still poorly understood. I first introduce into the general topic of delay systems and subsequently elaborate the relevant developments in machining applications, such as turning and milling. Finally I report on our own recent results, which reach from applications in machining to fundamental aspects of systems with time-varying delay.