Thermoelectric generators convert heat flows directly into usable electrical energy. In an appealing future scenario of application, waste heat could be partly recovered by thermoelectric genera tors. lt is typically argued that the heat comes without additional costs. Hence, every contribution of a thermoelectric generator may result in an increase of the over-all process efficiency.
A thermoelectric converter material with sufficient performance at competitive costs is hereby one precondition. A pragmatic way to optimize the thermoelectric performance of a material is the introduction of nanostructural elements into bulk materials. Hereby, the availability of high quality nanopowder fosters the development of novel thermoelectric materials: Materials which have per se an insufficient thermoelectric figure of merit can be optimized in a range that they become interesting substitute ma terials like nanocrystalline silicon or composites of nanocrystalline silicon and metal silicides.
High hot side temperatures require innovative device concepts. Same years ago, a thermoelectric generator concept was pro posed, which uses a large area pn junction aligned in parallel to the temperature gradient. This device concept is especially de signed for high hot side temperatures and large Ts of at least a few hundreds of degree C. All electrical and mechanical contacts on the pn junction device are on the cold side which is beneficial with respect to stability against degradation of the device. Using nanocrystalline silicon as thermoelectric converter material, we have shown that for temperature differences larger than 300°C, the pn junction device performs competitively with an ideal tradi tional device and outperforms this device if non-ideal contact re sistances of a real device are included.
Am 11. November 2015 kommen potentielle Arbeitgeber zum Campus Duisburg. Neben Informationsständen klären Firmen und Verbände in Workshops zu wichtigen Themen rund um den Berufseinstieg auf. Es gibt u.a. einen Bewerbungsmappen-Check und Impulsvorträge erfahrener Personaler, die den Besucherinnen und Besucher auch für Fragen zur Verfügung stehen. Einen Überblick über die diesjährigen Aussteller können Sie sich hier verschaffen:
We present new methods for solving systems of hyperbolic conservation laws with physical diffusion terms. In particular we are interested in solving the compressible Navier-Stokes equations. Our methods are extensions of a spacetime discontinuous Galerkin method for solving hyperbolic conservation laws [1].
In our extension to non-linear convection-diffusion equations we mostly follow the original scheme for the treatment of the non-linear terms: we use entropy variables as degrees of freedom and use entropy stable numerical fluxes. For incorporating the diffusion terms, we consider two different approaches: (i) the interior penalty method resulting in the extensions ST-NIPG and ST-SIPG, and
(ii) the local discontinuous Galerkin method resulting in the extension ST-LDG.
To guarantee stability of the methods both for fine and coarse grids, we also adjust the shock capturing terms of the original method appropriately to account for the presence of the diffusion term [2].
In this talk, we first give a short summary of the original method [1]. In the main part of the talk, we will present our new extensions and discuss their entropy stability properties -- followed by a numerical comparison of the new methods in one dimension. In particular, we will show numerical results for the compressible Navier-Stokes equations. We will conclude with thoughts about the extension to two dimensions.
[1] A. Hiltebrand and S. Mishra, Entropy stable shock capturing space-time discontinuous Galerkin schemes for systems of conservation laws. Numerische Mathematik, 126(1), 103--151, 2014.
[2] S. May, Spacetime discontinuous Galerkin methods for solving convection-diffusion systems, SAM report, 2015-05, 2015.
Kontakt:
Abteilung Bauwissenschaften, Institut für Mechanik
Informationsveranstaltung für:
Studierende der Ingenieurwissenschaften zu Studium und Praktikum im Ausland
- Kooperationen und Auslandskontakte
- Finanzierung und Stipendien
- Erfahrungsberichte und Eindrücke
Referenten:
Ira Terwyen (Akademisches Auslandsamt)
Prof. Rainer Leisten (Auslandsaufenthalt in Indien, Erasmus) Prof.in Nicole Krämer (Auslandsaufenthalt in den USA)
Dr. Stefan Werner (Kooperationen mit Südostasien)
Dr. Marc-André Weber (Auslandsaufenthalt in Brasilien)
Im Anschluss:
Gesprächsmöglichkeit mit Mitarbeitern und Studierenden über deren Auslandserfahrungen im Foyer SG 135
Veranstaltende:
SCIES Support Center for (International) Engineering Students
Entwicklung eines innovativen Lösungsansatzes für eine anwenderorientierte Prozessüberwachung und Optimierung von Medienführungskanälen beim Laser-Strahlschmelzen
The use of advanced software technologies is playing a central role in the process that leads to the automation of computational modeling. The problem of automation of computational methods has been explored by researches from the fields of mathematics, computer science and computational mechanics, resulting in a variety of approaches (e.g. a hybrid objectoriented approach and a hybrid symbolic-numeric approach) and available software tools (e.g. computer algebra systems, automatic differentiation tools, problem solving environments and numerical libraries). However, the true advantages of automation become apparent only if the description of the problem, the notation and the mathematical apparatus used are changed as well. One of the most evident examples of that is a new approach to the evaluation of matrix functions operating over tensors that are essential part of formulation of complex nonlinear material models in mechanics of solids. A method will be presented how to automatically derive numerically efficient closed-form representation of an arbitrary matrix function and its first and second derivatives for 3*3 matrices with real eigenvalues. Let us assume that there exists a numerical library of subroutines that numerically efficiently evaluates and returns selected matrix function and its first and second derivatives, calculated with machine precision accuracy on its whole definition interval. Then the matrix functions can be used in the same way as their scalar counterparts or any other elementary function.
Consequently, the formulation and numerical implementation of complex non-linear material models where the use of matrix functions is essential can be greatly simplified. This will be demonstrated on several examples from finite strain elasto-plasticity, Cam-Clay model, multiscale FE2 method etc. All the examples were made using the Mathematica based automatic code generation system AceGen (AceGen is available at http://symech.fgg.uni-lj.si).
