Refine
Document Type
- Conference Proceeding (76)
- Doctoral Thesis (17)
- Article (12)
- Master's Thesis (3)
- Periodical (2)
- Study Thesis (2)
- Diploma Thesis (1)
- Report (1)
Institute
- In Zusammenarbeit mit der Bauhaus-Universität Weimar (50)
- Institut für Strukturmechanik (ISM) (17)
- Graduiertenkolleg 1462 (12)
- Professur Angewandte Mathematik (4)
- Junior-Professur Augmented Reality (3)
- Professur Informatik im Bauwesen (3)
- Professur Stochastik und Optimierung (3)
- An-Institute (2)
- Institut für Europäische Urbanistik (2)
- Institut für Konstruktiven Ingenieurbau (IKI) (2)
Keywords
Year of publication
- 2010 (114) (remove)
The application of a recent method using formal power series is proposed. It is based on a new representation for solutions of Sturm-Liouville equations. This method is used to calculate the transmittance and reflectance coefficients of finite inhomogeneous layers with high accuracy and efficiency. Tailoring the refraction index profile defining the inhomogeneous media it is possible to develop very important applications such as optical filters. A number of profiles were evaluated and then some of them selected in order to perform an improvement of their characteristics via the modification of their profiles.
In the context of finite element model updating using output-only vibration test data, natural frequencies and mode shapes are used as validation criteria. Consequently, the correct pairing of experimentally obtained and numerically derived natural frequencies and mode shapes is important. In many cases, only limited spatial information is available and noise is present in the measurements. Therefore, the automatic selection of the most likely numerical mode shape corresponding to a particular experimentally identified mode shape can be a difficult task. The most common criterion for indicating corresponding mode shapes is the modal assurance criterion. Unfortunately, this criterion fails in certain cases and is not reliable for automatic approaches. In this paper, the purely mathematical modal assurance criterion will be enhanced by additional physical information from the numerical model in terms of modal strain energies. A numerical example and a benchmark study with experimental data are presented to show the advantages of the proposed energy-based criterion in comparison to the traditional modal assurance criterion.
In the context of finite element model updating using vibration test data, natural frequencies and mode shapes are used as validation criteria. Consequently, the order of natural frequencies and mode shapes is important. As only limited spatial information is available and noise is present in the measurements, the automatic selection of the most likely numerical mode shape corresponding to a measured mode shape is a difficult task. The most common criterion to indicate corresponding mode shapes is the modal assurance criterion. Unfortunately, this criterion fails in certain cases. In this paper, the pure mathematical modal assurance criterion will be enhanced by additional physical information of the numerical model in terms of modal strain energies. A numerical example and a benchmark study with real measured data are presented to show the advantages of the enhanced energy based criterion in comparison to the traditional modal assurance criterion.
Die vorliegende Arbeit beschäftigt sich mit der geometrischen Suffosionsbeständigkeit von Erdstoffen. Mit dem wahrscheinlichkeitstheoretischen Ansatz der Perkolationstheorie wurde ein analytisches Verfahren gewählt, mit dem suffosive Materialtransportprozesse modelliert und quantifiziert werden können. Mit dem verwendeten Perkolationsmodell wurde eine beliebige Porenstruktur eines realen Erdstoffes im 3-Dimensionalen modelliert. Mögliche Materialtransportprozesse innerhalb der modellierten Porenstruktur wurden anschließend simuliert. Allgemein gültige Gesetzmäßigkeiten wurden hergeleitet und Grenzbedingungen formuliert. Diese sind vom Erdstoff unabhängig und beschreiben Zusammenhänge zwischen Materialtransport und Porenstruktur. Anwendbar sind diese Ergebnisse auf homogene, isotrope und selbstähnliche Erdstoffgefüge. Aussagen über konkrete Erdstoffe können über die Transformationsmethode erfolgen. Für die Verwendung der Transformationsmethode ist vorab die relevante Porenstruktur, d. h. die Porenengstellenverteilung, zu ermitteln.
Visually impaired is a common problem for human life in the world wide. The projector-based AR technique has ability to change appearance of real object, and it can help to improve visibility for visually impaired. We propose a new framework for the appearance enhancement with the projector camera system that employed model predictive controller. This framework enables arbitrary image processing such as photo-retouch software in the real world and it helps to improve visibility for visually impaired. In this article, we show the appearance enhancement result of Peli's method and Wolffshon's method for the low vision, Jefferson's method for color vision deficiencies. Through experiment results, the potential of our method to enhance the appearance for visually impaired was confirmed as same as appearance enhancement for the digital image and television viewing.
Geotechnical constructions are sophisticated structures due to the non-linear soil behaviour and the complex soil-structure interaction, which entails great exigencies on the liable engineer during the design process. The process can be schematised as a difficult and, depending on the opportunities and skills of the processor more or less innovative, creative and heuristic search for one or a multiple of defined objectives under given boundary conditions. Wholistic approaches including numerical optimisation which support the constructing engineer in this task do not currently exist. Abstract problem formulation is not state of the art; commonly parameter studies are bounded by computational effort. Thereby potential regarding cost effectiveness, construction time, load capacity and/or serviceability are often used insufficiently. This paper describes systematic approaches for comprehensive optimisation of selected geotechnical constructions like combined pile raft foundations and quay wall structures. Several optimisation paradigms like the mono- and the multi-objective optimisation are demonstrated and their use for a more efficient design concerning various intentions is shown in example. The optimisation is implemented by using Evolutionary Algorithms. The applicability to geotechnical real world problems including nonlinearities, discontinuities and multi-modalities is shown. The routines are adapted to common problems and coupled with conventional analysis procedures as well as with numerical calculation software based on the finite element method. Numerical optimisation of geotechnical design using efficient algorithms is able to deliver highly effective solutions after investing more effort into the parameterization of the problem. Obtained results can be used for realizing different constructions near the stability limit, visualizing the sensitivity regarding the construction parameters or simply procuring more effective solutions.
Models in the context of engineering can be classified in process based and data based models. Whereas the process based model describes the problem by an explicit formulation, the data based model is often used, where no such mapping can be found due to the high complexity of the problem. Artificial Neuronal Networks (ANN) is a data based model, which is able to “learn“ a mapping from a set of training patterns. This paper deals with the application of ANN in time dependent bathymetric models. A bathymetric model is a geometric representation of the sea bed. Typically, a bathymetry is been measured and afterwards described by a finite set of measured data. Measuring at different time steps leads to a time dependent bathymetric model. To obtain a continuous surface, the measured data has to be interpolated by some interpolation method. Unlike the explicitly given interpolation methods, the presented time dependent bathymetric model using an ANN trains the approximated surface in space and time in an implicit way. The ANN is trained by topographic measured data, which consists of the location (x,y) and time t. In other words the ANN is trained to reproduce the mapping h = f(x,y,t) and afterwards it is able to approximate the topographic height for a given location and date. In a further step, this model is extended to take meteorological parameters into account. This leads to a model of more predictive character.
Seit 1969 werden für die Bundesrepublik kontinuierlich Berechnungen zu den Gesamtkosten des Straßenverkehrs der Bundesfernstraßen und deren Verteilung auf die Verkehrsteilnehmer durchgeführt. Die Ergebnisse der Wegekostenrechnungen der Jahre 2002 und 2007 sind die Grundlage für die mittlerweile für das deutsche Autobahnnetz eingeführte fahrleistungsbezogene Benutzungsgebühr für Lkw mit einem zulässigen Gesamtgewicht von mindestens zwölf Tonnen. Damit wird die Forderung der EU-Richtlinie 1999/62/EG umgesetzt, nach der sich die durchschnittlichen Straßenbenutzungsgebühren an den Kosten für den Bau, den Betrieb und den Ausbau des betreffenden Verkehrswegenetzes orientieren sollen. Mit der EU-Richtlinie 2006/38/EG kündigt sich die weitere Entwicklung bei der Berechnung von Straßenbenutzungsgebühren an. Zukünftig sollen auch externe Kosten in die Berechnung einfließen. Ein erster Schritt zur Berücksichtigung dieser externen Kosten erfolgte mit Erstellung eines Handbuchs im Rahmen eines EU-Forschungsprojektes. Das Handbuch enthält aufgrund der unterschiedlichen Rahmenbedingungen in den Mitgliedsstaaten der EU keine exakten Berechnungsvorschriften, sondern stellt verschiedene methodische Ansätze bisher durchgeführter Studien zu externen Kosten vor, gibt Empfehlungen hinsichtlich der Methodenwahl und beinhaltet Schätzungen über die Höhe der externen Kosten. Die im europäischen Raum in den vergangenen Jahren durchgeführten Studien zur Ermittlung externer Kosten des Verkehrs zeichnen sich durch einander ähnelnde Vorgehensweisen aus, die aber vor allem hinsichtlich der Kostenrechnungsart und der verwendeten Kostensätze aus Sicht des Verfassers der vorliegenden Arbeit kritische Aspekte aufweisen. In der vorliegenden Dissertationsschrift wird daher eine alternative Berechnungsmethodik zur Ermittlung abschnitts-, fahrzeugklassen- und fahrleistungsbezogener externer Kosten für Autobahnen entwickelt und an einem ausgewählten Beispielnetz zur Anwendung gebracht. Dabei wird in einigen wesentlichen Punkten von der in aktuellen Studien überwiegend gewählten Vorgehensweise abgewichen, um eine andere Sichtweise darzustellen. Damit trägt die vorliegende Arbeit substanziell zur Erweiterung des Erkenntnisstands zu Berechnungsmethoden externer Kosten des Straßenverkehrs bei. Die hier entwickelte Berechnungsmethodik ist außerdem als Grundlage für ein in der Praxis anwendbares Verfahren zu verstehen und zeichnet sich auch daher durch eine einfach zu handhabende Übertragbarkeit auf das gesamte Autobahnnetz Deutschlands aus. Die Abschnitte entsprechen den Teilstrecken zwischen zwei Autobahnanschlussstellen. Es wird zwischen den beiden Fahrzeugklassen "Lkw ab 12 t zulässigem Gesamtgewicht" und "Sonstigen Fahrzeugen" unterschieden. Obwohl momentan nur eine Benutzungsgebühr für Lkw ab 12 t zulässigem Gesamtgewicht erhoben wird, ist es mit der entwickelten Methodik möglich, fahrleistungsbezogene externe Kosten für alle Kfz angeben zu können. Die Einbeziehung externer Nutzen wird in diesem Zusammenhang andiskutiert; der Schwerpunkt liegt allerdings auf den externen Kosten. Im Rahmen der Arbeit werden zunächst Definitionen wesentlicher Terminologien dargestellt, soweit diese für das Verständnis der sich anschließenden Diskussion und Festlegung der Grundlagen der entwickelten Berechnungsmethodik notwendig erscheinen. Diese Diskussion und Festlegung umfasst die Bereiche Kostenrechnungsart, Bewertungsverfahren zur Ermittlung des Wertegerüsts, Diskontrate, zu betrachtende Kostenbereiche, Mengengerüst und Allokationsrechnung. Darauf folgend werden die betrachteten Kostenbereiche anhand vorliegender Studien und eigener Überlegungen detailliert dargestellt und das Wertegerüst bestimmt. Außerdem wird die Allokationsrechnung und das für die Berechnung heranzuziehende Mengengerüst für jeden Bereich separat vorgestellt. Anschließend wird die entwickelte Berechnungsmethodik auf ein Beispielnetz (Autobahnnetz Thüringen) angewendet. Neben der Vorstellung des Untersuchungsgebiets, der Berechnung der externen Kosten und der disaggregierten Ergebnisdarstellung wird die Einteilung des Beispielnetzes in unterschiedliche Preiskategorien auf der Grundlage der abschnittsbezogen vorliegenden Ergebnisse diskutiert, auf deren Basis die externen Kosten über Straßenbenutzungsgebühren internalisiert werden könnten. Im Rahmen einer Sensitivitätsanalyse werden einzelne Annahmen der Berechnungsmethodik bzw. Kostensätze des Wertegerüsts variiert. Die Auswirkungen dieser Variationen werden wiederum am Beispielnetz, für das erneute Kostenberechnungen vorgenommen werden, dargelegt. Abschließend werden offen gebliebene Fragestellungen und Empfehlungen für weitere Untersuchungen benannt.
The changed global security situation in the last eight years has shown the importance of emergency management plans in public buildings. Therefore, the use of computer simulators for surveying fire safety design and evacuation process is increasing. The aim of these simulators is to have more realistic evacuation simulations. The challenge is, firstly, to realize the virtual simulation environment based on geometrical and material boundary conditions, secondly, to considerate the mutual interaction effects between different parameters and, finally, to have a realistic visualization of the simulated results. In order to carry out this task, an especial new software method on a BIM-platform has to be developed which can integrate all required simulations and will be able to have an immersive output BIM ISEE (Immersive Safety Engineering Environment). The new BIM-ISEE will integrate the Fire Dynamics Simulator (FDS) for fire and evacuation simulation in the Autodesk Revit which is a BIM-platform and will represent the simulation results in the immersive virtual environment at the institute (CES-Lab). With BIM-ISEE the fire safety engineer will be able to obtain more realistic visualizations in the immersive environment, to modify his concept more effectively, to evaluate the simulation results more accurately and to visualize the various simulation results. It can also give the rescue staff the opportunity to perform and evaluate emergency evacuation trainings.
Die Planung von komplexen Bauwerken erfolgt zunehmend mit Planungswerkzeugen, die den Export von Bauwerksinformationen im STEP-Format auf Grundlage der IFC (Industry Foundation Classes) erlauben. Durch die Verfügbarkeit dieser Schnittstelle ist es möglich, Bauwerksinformationen für die weiterführende Verarbeitung zu verwenden. Zur Visualisierung der geometrischen Daten stehen innerhalb der IFC verschiedene geometrische Modelle für die Darstellung von Bauteilen zur Verfügung. Unter anderem werden für das „Ausschneiden“ von Öffnungen aus Bauteilen (z.B. für Fenster und Türen) geometrische boolesche Operationen benötigt.
Gegenstand des Beitrags ist die Vorstellung eines Algorithmus zur Berechnung von booleschen Operationen auf Basis eines triangulierten B-Rep (Boundary Representation) Modells nach HUBBARD (1990). Da innerhalb von IFC-Gebäudemodellen Bauteile oft das Resultat mehrerer boolescher Operationen sind (z.B. um mehrere Fensteröffnungen von einer gegebenen Wand abzuziehen), wurde der Algorithmus von Hubbard angepasst, sodass mehrere boolesche Operationen gleichzeitig berechnet werden können. Durch diese Optimierung wird eine deutliche Reduzierung der benötigten Berechnungen und somit der Rechenzeit erreicht.
The main aim of the research project in progress is to develop virtual models as tools to support decision-making in the planning of construction maintenance. The virtual models gives the capacity to allow them to transmit, visually and interactively, information related to the physical behaviour of materials, components of given infrastructures, defined as a function of the time variable. The interactive application allows decisions to be made on conception options in the definition of plans for maintenance, conservation or rehabilitation. The first virtual prototype that is now in progress concerns just lamps. It allows the examination of the physical model, visualizing, for each element modelled in 3D and linked to a database, the corresponding technical information concerned with the wear and tear aspects of the material, calculated for that period of time. In addition, the analysis of solutions for repair work or substitution and inherent cost are predicted, the results being obtained interactively and visualized in the virtual environment itself. The aim is that the virtual model should be able to be applied directly over the 3D models of new constructions, in situations of rehabilitation. The practical usage of these models is directed, then, towards supporting decision-making in the conception phase and the planning of maintenance. In further work other components will be analysed and incorporated into the virtual system.
Tests on Polymer Modified Cement Concrete (PCC) have shown significant large creep deformation. The reasons for that as well as additional material phenomena are explained in the following paper. Existing creep models developed for standard concrete are studied to determine the time-dependent deformations of PCC. These models are: model B3 by Bažant and Bajewa, the models according to Model Code 90 and ACI 209 as well as model GL2000 by Gardner and Lockman. The calculated creep strains are compared to existing experimental data of PCC and the differences are pointed out. Furthermore, an optimization of the model parameters is performed to fit the models to the experimental data to achieve a better model prognosis.
Virtual reality systems offer substantial potential in supporting decision processes based purely on computer-based representations and simulations. The automotive industry is a prime application domain for such technology, since almost all product parts are available as three-dimensional models. The consideration of ergonomic aspects during assembly tasks, the evaluation of humanmachine interfaces in the car interior, design decision meetings as well as customer presentations serve as but a few examples, wherein the benefit of virtual reality technology is obvious. All these tasks require the involvement of a group of people with different expertises. However, current stereoscopic display systems only provide correct 3D-images for a single user, while other users see a more or less distorted virtual model. This is a major reason why these systems still face limited acceptance in the automotive industry. They need to be operated by experts, who have an advanced understanding of the particular interaction techniques and are aware of the limitations and shortcomings of virtual reality technology. The central idea of this thesis is to investigate the utility of stereoscopic multi-user systems for various stages of the car development process. Such systems provide multiple users with individual and perspectively correct stereoscopic images, which are key features and serve as the premise for the appropriate support of collaborative group processes. The focus of the research is on questions related to various aspects of collaboration in multi-viewer systems such as verbal communication, deictic reference, embodiments and collaborative interaction techniques. The results of this endeavor provide scientific evidence that multi-viewer systems improve the usability of VR-applications for various automotive scenarios, wherein co-located group discussions are necessary. The thesis identifies and discusses the requirements for these scenarios as well as the limitations of applying multi-viewer technology in this context. A particularly important gesture in real-world group discussions is referencing an object by pointing with the hand and the accuracy which can be expected in VR is made evident. A novel two-user seating buck is introduced for the evaluation of ergonomics in a car interior and the requirements on avatar representations for users sitting in a car are identified. Collaborative assembly tasks require high precision. The novel concept of a two-user prop significantly increases the quality of such a simulation in a virtual environment and allows ergonomists to study the strain on workers during an assembly sequence. These findings contribute toward an increased acceptance of VR-technology for collaborative development meetings in the automotive industry and other domains.
In order to model and simulate collapses of large scale complex structures, a user-friendly and high performance software system is essential. Because a large number of simulation experiments have to be performed, therefore, next to an appropriate simulation model and high performance computing, efficient interactive control and visualization capabilities of model parameters and simulation results are crucial. To this respect, this contribution is concerned with advancements of the software system CADCE (Computer Aided Demolition using Controlled Explosives) that is extended under particular consideration of computational steering concepts. Thereby, focus is placed on problems and solutions for the collapse simulation of real world large scale complex structures. The simulation model applied is based on a multilevel approach embedding finite element models on a local as well as a near field length scale, and multibody models on a global scale. Within the global level simulation, relevant effects of the local and the near field scale, such as fracture and failure processes of the reinforced concrete parts, are approximated by means of tailor-made multibody subsystems. These subsystems employ force elements representing nonlinear material characteristics in terms of force/displacement relationships that, in advance, are determined by finite element analysis. In particular, enhancements concerning the efficiency of the multibody model and improvements of the user interaction are presented that are crucial for the capability of the computational steering. Some scenarios of collapse simulations of real world large scale structures demonstrate the implementation of the above mentioned approaches within the computational steering.
CONSTITUTIVE MODELS FOR SUBSOIL IN THE CONTEXT OF STRUCTURAL ANALYSIS IN CONSTRUCTION ENGINEERING
(2010)
Parameters of constitutive models are obtained generally comparing the results of forward numerical simulations to measurement data. Mostly the parameter values are varied by trial-and-error in order to reach an improved fit and obtain plausible results. However, the description of complex soil behavior requires advanced constitutive models where the rising complexity of these models mainly increases the number of unknown constitutive parameters. Thus an efficient identification "by hand" becomes quite difficult for most practical geotechnical problems. The main focus of this article is on finding a vector of parameters in a given search space which minimizes discrepancy between measurements and the associated numerical result. Classically, the parameter values are estimated from laboratory tests on small samples (triaxial tests or oedometer tests). For this purpose an automatic population-based approach is present to determine the material parameters for reconstituted and natural Bothkennar Clay. After the identification a statistical assessment is carried out of numerical results to evaluate different constitutive models. On the other side a geotechnical problem, stone columns under an embankment, is treated in a well instrumented field trial in Klagenfurt, Austria. For the identification purpose there are measurements from multilevel-piezometers, multilevel-extensometers and horizontal inclinometer. Based on the simulation of the stone columns in a FE-Model the identification of the constitutive parameters is similar to the experimental tests by minimizing the absolute error between measurement and numerical curves.
Buildings can be divided into various types and described by a huge number of parameters. Within the life cycle of a building, especially during the design and construction phases, a lot of engineers with different points of view, proprietary applications and data formats are involved. The collaboration of all participating engineers is characterised by a high amount of communication. Due to these aspects, a homogeneous building model for all engineers is not feasible. The status quo of civil engineering is the segmentation of the complete model into partial models. Currently, the interdependencies of these partial models are not in the focus of available engineering solutions. This paper addresses the problem of coupling partial models in civil engineering. According to the state-of-the-art, applications and partial models are formulated by the object-oriented method. Although this method solves basic communication problems like subclass coupling directly it was found that many relevant coupling problems remain to be solved. Therefore, it is necessary to analyse and classify the relevant coupling types in building modelling. Coupling in computer science refers to the relationship between modules and their mutual interaction and can be divided into different coupling types. The coupling types differ on the degree by which the coupled modules rely upon each other. This is exemplified by a general reference example from civil engineering. A uniform formulation of coupling patterns is described analogously to design patterns, which are a common methodology in software engineering. Design patterns are templates for describing a general reusable solution to a commonly occurring problem. A template is independent of the programming language and the operating system. These coupling patterns are selected according to the specific problems of building modelling. A specific meta-model for coupling problems in civil engineering is introduced. In our meta-model the coupling patterns are a semantic description of a specific coupling design.
CRITICAL STRESS ASSESSMENT IN ANGLE TO GUSSET PLATE BOLTED CONNECTION BY SIMPLIFIED FEM MODELLING
(2010)
Simplified modelling of friction grip bolted connections of steel member – to – gusset plate is often applied in engineering practise. The paper deals with the simplification of pre-tensioned bolt model and simplification of load transfer within connection. Influence on normal strain (and thus stress) distribution at critical cross-section is investigated. Laboratory testing of single-angle or double-angle members – to – gusset plates bolted connections were taken as basis for numerical analysis. FE models were created using 1D and 2D elements. Angles and gusset plates were modelled with shell elements. Two methods of modelling of friction grip bolting were considered: bolt-regarding approach with 1D element systems modelling bolts and two variants of bolt-disregarding approach with special constraints over some part of member and gusset plate surfaces in contact: a) constraints over whole area of contact, b) constraints over the area around each bolt shank (“partially tied”). Modelling of friction grip bolted connections using simplified bolt modelling may be effective, especially in the case of analysis concerning elastic range only. In such a case disregarding bolts and replacing them with “partially tied” modelling seems to be more attractive. It is less time-consuming and provides results of similar accuracy in comparison to analysis utilizing simplified bolt modelling.
This thesis focuses on the cryptanalysis and the design of block ciphers and hash func- tions. The thesis starts with an overview of methods for cryptanalysis of block ciphers which are based on differential cryptanalysis. We explain these concepts and also sev- eral combinations of these attacks. We propose new attacks on reduced versions of ARIA and AES. Furthermore, we analyze the strength of the internal block ciphers of hash functions. We propose the first attacks that break the internal block ciphers of Tiger, HAS-160, and a reduced round version of SHACAL-2. The last part of the thesis is concerned with the analysis and the design of cryptographic hash functions. We adopt a block cipher attack called slide attack into the scenario of hash function cryptanalysis. We then use this new method to attack different variants of GRINDAHL and RADIOGATUN. Finally, we propose a new hash function called TWISTER which was designed and pro- posed for the SHA-3 competition. TWISTER was accepted for round one of this com- petition. Our approach follows a new strategy to design a cryptographic hash function. We also describe several attacks on TWISTER and discuss the security issues concern- ing these attack on TWISTER.