Program - Efteråret 2013

Seminarerne afholdes (hvor andet ikke er oplyst) hver onsdag i bygn 27.1, lokale I og varer fra kl. 13.00 til kl. 15.00.  Arrangør: Martin Niss

 - Se tidligere seminar programmer her

 
04.09 Common seminar about the mathematics and physics eductions arranged by the study boards

11.09 Arnold Skimminge
Neuroscience from a physicists perspective

Physics and mathematics have become an integral part of neuroscience in the past decades, primarily due to technological advances in imaging. I will share my personal experiences within the neuroscientific research field as a physicist, and give some insights into the study of the human brain. My experiences has primarily been with magnetic resonance imaging, which rely heavily on physics to optimize the scanners to observe specific dynamics, as well as advanced mathematics to give interpretable and quantified results.

18.09. Adam Mahdi (Department of Mathematics, North Carolina State University)

Algebraic methods in cardiovascular modeling

Understanding the cardiovascular (CV) control system is crucial in order to detect pathologies. The main role of the CV system is to maintain adequate oxygenation of all tissues. This is achieved by maintaining blood flow and pressure at a fairly constant level. An important contributor to the short-term CV control is the baroreflex, which uses specialized neurons, called baroreceptors, that are activated using mechanosensitive sensors located in the aortic arch and carotid sinuses. In this talk we will discuss various mathematical techniques and approaches  to construct a biologically motivated and computationally efficient model of baroreceptor activity. Moreover, we will also consider the problem of  structural identifiability restricted to the important class of viscoelastic mechanical systems used in the CV modeling. In particular, we show how algebraic methods can be used to design a structurally identifiable model.

This work is a part of the bigger project called "The Virtual Physiological Rat Project" (http://virtualrat.org/), with objective to build and simulate the cardiovascular functions of the rat and build a validated computer models across rat strains. The ultimate goal of using mathematical and computer models is to predict the physiological characteristics of not yet realized combinations, derive those combinations in the lab, and then test the predictions.

25.09. No seminar (Nat-dag)

02.10. Problem formulation seminar arranged by the mathematics and physics study boards

09.10. Det 26. Heldagsseminar om matematikkens og fysikkens fagdidaktik (kl. 9.30 - 16.00)
Sammenhæng og udvikling i matematikundervisningen i grundskole og gymnasium

16.10. No seminar (fall break)

23.10. Michael May (Institut for naturfagenes didaktik, KU)

Student’s graph and model comprehension problems in science and engineering education:  Semiotic perspectives

A long-term interest of mine is the development of a semiotics of modeling in science and engineering education. Semiotics is the study of meaning and signification from the point of view of a signs and sign systems. This is a broad and complex issue and I will say something about what a “semiotics of modeling” might be at the end of my talk, but I will mainly focus on a particular subset of conceptual problems that appear across different domains of university science education and which seem to be related to “graph comprehension”. I will start by presenting a number of concrete examples of “misconceptions” (from conceptual tests and/or interviews with students) and then provide some hypothetical explanations for why these problems occur. Teachers in science and engineering often complain about student’s mathematical competences, but one of the difficult issues that we might take up in discussion is how to entangle mathematical competence problems from disciplinary conceptual problems in understanding the underlying concepts and models. Graphs interpreted by or produced by students gives access to these problems, and we should discuss what didactic consequences we might draw from considering such recurrent problems of graph- and model-comprehension.

30.10. Lærke Bang Jaocbsen (Borupgaard Gymnasium)

Gymnasiet tænkt forfra - set fra en deltagende matematik- og fysiklærer

Region Hovedstaden har i starten af 2012 iværksat projektet Gymnasiet tænkt forfra. Projektet bygger ovenpå forprojektet Innovationskraft og Entreprenørskab i Gymnasieskolen, hvor udkommet bl.a. var at inden for gymnasieskolens strukturer og kulturer er det svært at arbejde mod øgede innovative kompetencer for gymnasieeleverne.

Projektet Gymnasiet tænkt forfra består af seks skoleklasser på tværs af gymnasier, hvor fem lærere på hver skole arbejder sammen om at øge elevernes innovative kompetencer. I oplægget vil matematik- og fysiklæreren på den deltagende skole Borupgaard Gymnasium fortælle om udgangspunktet for projektet, hvordan projektet er blevet grebet an, og hvordan eleverne har taget imod projektet. De spæde forskningsresultater fra følgeforskningen vil blive præsenteret.

06.11. Jacob Schach Møller

Anden halvdel af Hilberts 16. problem

13.11. Anne Elisabeth Sejten (Institut for kultur og identitet, RUC)
Algebra, geometri og æstetik hos Paul Valéry

Takket være en omfattende forfattervirksomhed som lejlighedsskribent og essayist fremstår den franske digter Paul Valéry (1871-1945) som en vigtig kilde til vor tids forståelse af kunst og æstetik. Et tankevækkende – og måske overraskende – aspekt ved hans æstetiske tænkning er en fascination af eksakt videnskab generelt og geometrien i særdeleshed. Associationer til algebra, matematik og geometri inspirerer den æstetiske refleksion. Det gælder tekster om både poesi (Stéphane Mallarmé), dans, malerkunst (Leonardo da Vinci), arkitektur og filosofi (Descartes). Foredraget vil fremdrage og belyse disse forbindelseslinjer mellem æstetik og geometri og således pege på, hvordan det geometriske sprog kan afstedkomme videnskabsteoretiske overvejelser i et humanvidenskabeligt perspektiv.

20.11. Dorthe Posselt (IMFUFA)

Structure and Dynamics of Biological Membranes

27.11. Claire Louise Vincent (Department of Wind Energy, DTU)

Modelling the wind for wind energy resource assessment and forecasting

Accurate predictions of the wind speed are important in all aspects of wind farm planning and operation. In regions where long term measurements are not available, or where wind resource assessment is required over a large spatial area, mesoscale modelling combined with statistical or dynamical downscaling can often provide a good estimate of the wind field. A mesoscale model solves the equations of motion for the atmosphere on a grid with a horizontal grid spacing of around 1-12 km, combined with parametrizations for sub-grid-scale physical processes such as vertical mixing and surface layer properties.

In this presentation, I will give an introduction to mesoscale models, and describe some of the ways they are used for wind resource assessment at DTU wind energy. I will present some results of idealised experiments that we have carried out to assess the response of the model to a single change in surface roughness, and show why further downscaling of the mesocale model output is essential for wind resource assessment. I will discuss ways in which we verify the mesoscale model output, in both temporal and spectral space.

04.12. Claire Lemarchand (IMFUFA)
Computer simulations of COOEE bitumen: from chemical to mechanical properties

11.12. Helle Sørensen (Department of Mathematical Sciences, University of Copenhagen)

Functional data analysis – Tools and examples

Functional data are data that can be represented by suitable functions, such as curves or surfaces. Each function is viewed as a single sample element rather than as a collection of sample elements. I will present examples on functional data from different scientific fields, namely data concerning horse gait, composition of plants, and geometry of blood vessels. With those datasets in mind I will discuss some of the challenges that are characteristic for functional data. First, functions are in practice only recorded discretely and with noise, so smoothing is an integral part of the analysis. Second, even if curves from different subjects exhibit the same features, these features may occur at different times, so it is often appropriate to align the curves as part of the analysis. Third, one is often interested in associations between function data and other measurements, so we need regression type analyses for functional data.