None

The work deals with the task of forecasting motorway traffic states. The data of inductive loop detectors, placed on a large scale German motorway network, are analysed in regard to develop new traffic forecast algorithms. An application is presented that uses the new algorithms in operation for online traffic information.

To see this Video, you need the <a href="http://scopes.real.com/real/player/player.html?src=de_cp,rpchoiceintl_de,rpchoiceintl_de_1&lang=de&dc=829828827">RealPlayer 8 Basic-Version</a>. If you are a member of the Gerhard Mercator University of Duisburg, you may use <a href="http://sms-hrz.uni-duisburg.de:8081/install/rp8intra.exe">this version of the Realplayer</a>

This work presents an analysis of the energy structure of carriers enclosed in self-assembled InAs quantum dots. Capacitance-voltage-spectroscopy (CV) with perpendicular magnetic fields is used to identify the energy structure of holes in quantum dots. Applying a two-dimensional harmonic oscillator model, we demonstrate an irregular filling sequence where the d states are occupied with holes before the filling of the p shell is completed. This surprising behavior is explained by a shift in the energy structure in favor of the coulomb repulsion due to the strong interaction in the hole system. We have mapped out the wave functions of electron and hole carriers using CV-spectroscopy with parallel magnetic fields. This allows us to obtain two dimensional plots of the probability densities in k-space for the carriers. The wave functions are interpreted in a quasi particle picture. For the s electrons in quantum dots we obtain Gaussian like probability densities with certain deviations. The wave functions are elongated along [1-10] in direct space. This is caused by a morphological anisotropy of the quantum dots and due to the piezoelectric effect in these structures. The p states show nodal structures with orbitals that are oriented perpendicular to each other. The low energy p states are oriented along [1-10] in direct space. The hole wave functions show an elongation along the perpendicular direction [110] in direct space. This confirms the assumption that the probability distribution is mainly influenced by piezoelectric effects in the strained semiconductor system. Using polarization dependent photoluminescence spectroscopy we demonstrate an energy shift in the interband transitions which indicates an anisotropic confining potential.

A detailed analysis of single-vehicle data is presented that sheds some light on the microscopic interaction of the vehicles in the various traffic states. Based on these results an improved cellular automaton model for traffic flow incorporating anticipation effects, reduced acceleration capabilities and an enhanced interaction horizon for braking is proposed. The model is able to reproduce the three phases (free flow, synchronized traffic, and wide jams) observed in real traffic. Furthermore a good agreement with detailed empirical single-vehicle data in all phases can be found. It turns out, that the incorporation of the human desire for smooth and comfortable driving into the driving strategy of vehicles leads to a model that exhibits synchronized traffic. Anticipation effects are responsible for a stabilization of the traffic phases and the empirically observed coexistence of wide moving jams with both free flow and synchronized traffic can be reproduced. It is shown that the single-lane dynamics can be extended to the two-lane case without changing the basic (realistic) properties of the model. Therefore it is possible to reproduce special two-lane phenomena, like the synchronization of the lanes, the lane usage inversion and the density-dependence of the number of lane changes, without adapting the parameters of the model. Finally, a statistical analysis of traffic data that is provided by an area-wide coverage of the highway network of North Rhine-Westfalia with inductive loops is given. The identification and characterization of the bottlenecks of the network reveal that the bottlenecks are not of topological nature but are constituted by onramps. This underscores the applicability of ramp metering systems.

Seit einigen Jahren gewinnt die Forschung an mehrkomponentigen heterogenen Polymersystemen zunehmend Bedeutung. Dies liegt neben dem akademischen Forschungsinteresse auch an dem grossen wirtschaftlichen Interesse in der industriellen Anwendung. In der Materialforschung werden aufgrund des gestiegenen Bedarfs an vielseitigen und preisgüunstigen Hochleistungswerkstoffen mit einem breiten Eigenschaftsprofil neue Polymerwerkstoffe durch Mischung vorhandener Homopolymere entwickelt. Von großer industrieller Bedeutung ist die Kombination der beiden Polymere isotaktisches Polypropylen (iPP) und Polyethylen (PE). In diesen Systemen wird iPP und PE in bestimmten Gewichtsverhältnissen gemischt, um die Eigenschaften des Werkstoffs gezielt zu verändern. Die Verbesserungen des Eigenschaftsprofils werden häufig als das Resultat synergetischer Effekte zwischen den jeweiligen Komponenten des Systems erklärt. Jedoch stellt das Polymersystem iPP/PE innerhalb des gesamten Konzentrationsbereichs ein nichtmischbares Polymersystem dar. Die resultierenden neuen Materialeigenschaften des Systems iPP/PE erreichen deshalb oft nicht das geforderte Eigenschaftsprofil. Aufgrund der Nichtmischbarkeit kommt es zu einer Phasenseparation. Diese führt u.a. zu einer inhomogenen Verteilung der Minorphase in der Matrix. Die schlechte Adhäsion an den Phasengrenzen beeinträchtigt die Kraftübertragung bei Belastung des Materials und beeinflußt so die mechanischen Eigenschaften des Systems. Eine Moeglichkeit, diese aus der Nichtmischbarkeit der Ausgangskomponenten resultierenden Nachteile zu vermeiden, besteht in dem Einsatz von grenzflaechenaktiven Polymeren, sogenannten Vertraeglichkeitsmachern Haeufig werden sogenannte Block- oder Pfropf-Copolymere als Vertraeglichkeitsmacher eingesetzt. Sie sollen in der Phasengrenze der nichtmischbaren Polymere derart lokalisiert sein, dass sie die Grenzphase ueberbruecken. Eine Vielzahl von Untersuchungen wurde durchgefuehrt, um fuer spezielle Polymersysteme die passenden Vertraeglichkeitsmacher zu entwickeln.

Magnetic models with strong disorder are the topic of this work. Diluted antiferromagnets in an external field (DAFF) are of special interest, both from a theoretical, and from an experimental point of view. In this work critical exponents at various points in the temperature vs. magnetic field phase diagram are determined by the analysis of exact calculated ground states and of Monte Carlo simulations. The specific heat and its critical exponent deserve special attention. Simulations of irreversibilities of the specific heat are in agreement with latest measurements. Further emphasis is put on the investigation of the non-equilibrium dynamics of the DAFF. The relaxation dynamics of the DAFF is governed by thermal activation. Further it is shown that also the DAFF exhibits Aging effects, as they are known from many other disordered systems

In this thesis several polymer materials have been investigated with respect to their application as optical chemo-sensors, which are based on molecule diffusion into thin sensitive layers. The dynamic response of such sensors is mainly controlled by the diffusion kinetics of the molecules in the film. The solution of the diffusion differential equation (2nd Ficks law) in the case of thin films yields the temporary concentration profile of the molecules in the film. Linearity between the refractive index changes and the concentration follows directly from the Lorentz-Lorenz equation and therefore gives the same variation for the index changes. With that the development of the refractive index profile during in- and out-diffusion is well described by the diffusion theory and consequently the dynamic response of the optical sensor can be modelled by a suitable theory. Thin glass/silver/polymer multilayer systems have been characterised with respect to their optical sensing parameters by using metal film enhanced leaky mode spectroscopy. This useful optical technique, which offers the observation of the surface plasmon resonance and the leaky modes of thin dielectric films, has been improved for the analysis of inhomogeneous refractive index profiles by using a transfer-matrix formalism for layered media. Furthermore, waveguide birefringence experiments on thin anisotropic polyimide films as planar lightguides have been carried out to show how the sensitivity of the sensor can be optimised by a suitable choice or a specific modification of the polymer material.

Anisotropies essentially affect magnetism in thin ferromagnetic films of few atomic layers. On the one hand they can stabilize long range order in these systems, on the other hand they strongly influence the orientation of magnetization. The intrinsic causes of anisotropies in these systems are the spin-orbit coupling of the electrons and the long-range magnetic dipole interaction. While the dipole interaction always favors an orientation of magnetization in the plane of the film, spin-orbit coupling can favor different orientations of magnetization on the surface and in the inner layers of the film. This can lead to a competition between anisotropies, which in turn leads to a spin reorientation transition with varying film thickness. This transition can be of varying order, i. e. be continuous or discontinuous. Experiments also find a spin reorientation transition with varying temperature, which until now was not well understood. In the framework of a classical Heisenberg model this transition is investigate d by means of different theoretical methods in the course of this thesis. At zero temperature the system can be dealt with analytically and criteria for the spin reorientation transition and its order are found. Furthermore it is investigated whether the long-range dipole interaction results in a domain ground state. These investigations are extended to finite temperatures by means of a molecular field theory and results are compared to Monte Carlo simulations. It is shown that in contrast to other works the temperature driven spin reorientation transition in the monolayer is discontinuous also in the simulations, whereas in general it is continuous for the bilayer. Consequently the molecular field theory and the Monte Carlo simulations agree qualitatively. Exemplary for thicker films the influence of an external magnetic field is investigated in the bilayer, furthermore the effective anisotropies Kn(T) of the phenomenological Landau theory are calculated numerically for the microscopic model. Analytic expres sions for the dependence of the anisotropies Kn(T) on the parameters of the model are obtained by the means of perturbation theory, which lead to a deeper understanding of the spin reorientation transition. Accordingly to this the origin for the spin reorientation transition lies in the differing temperature dependence of the dipolar and spin-orbit parts of the Kn(T). Additionally the magnetization in the surface of the film decreases more rapidly with increasing temperature. As a consequence the influence of the surface anisotropy decreases with increasing temperature. This effects a similar result as increasing the film thickness and leads to the transition. Finally the model is extended to continuous film thicknesses, since in experiment the spin reorientation transition depends crucially on film thickness. The results of this extended model are compared to experiment and give good agreement.

In this work potential benefits and drawbacks of providing information to road users are discussed. To model the complex behaviour of road users multi-agent-techniques are used. A special agent-architecture is proposed and discussed. To consider the impact of pre-trip-information, coordination games, like the El-Farol problem and the minority game, are studied and interpreted in the context of route choice scenarios. Additionally, the impact of en-route information is studied by simulating a basic two-route scenario with dynamic and static agents. A correlation analysis yields that the traffic system is destabilised by en-route information. The overall performance of the system is reduced by the effect of concentration. Different types and ways of generating current information are tested. It is found that the nature of the information influences the potential benefits of the information system strongly. As potential application an Advanced Traveller Information System is introduced, which comprises an agent-based simulator, which is coupled to a stream of on-line data. The framework provides network-wide information about the current traffic state. It is applied to the freeway network of the felderal state of North-Rhine Westfalia (NRW). Results are presented and further applications are discussed. To provide predictive information different methods to forecast traffic are analysed and discussed. Heuristics generated by an statistical analysis of historical data are studied. The daily traffic demand is classified with regard to daily, seasonal characteristics, as well as special events or directional flow. To provide a prediction for arbitrary horizons heuristics are merged with a constant model. The model delivers good predictions for a short and long-term traffic forecast.