Refine
Institute
- In Zusammenarbeit mit der Bauhaus-Universität Weimar (141) (remove)
Keywords
- Architektur <Informatik> (141) (remove)
The contribution presents a model that is able to simulate construction duration and cost for a building project. This model predicts set of expected project costs and duration schedule depending on input parameters such as production speed, scope of work, time schedule, bonding conditions and maximum and minimum deviations from scope of work and production speed. The simulation model is able to calculate, on the basis of input level of probability, the adequate construction cost and time duration of a project. The reciprocal view attends to finding out the adequate level of probability for construction cost and activity durations. Among interpretive outputs of the application software belongs the compilation of a presumed dynamic progress chart. This progress chart represents the expected scenario of development of a building project with the mapping of potential time dislocations for particular activities. The calculation of a presumed dynamic progress chart is based on an algorithm, which calculates mean values as a partial result of the simulated building project. Construction cost and time models are, in many ways, useful tools in project management. Clients are able to make proper decisions about the time and cost schedules of their investments. Consequently, building contractors are able to schedule predicted project cost and duration before any decision is finalized.
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.
Objects for civil engineering applications can be identified with their reference in memory, their alpha-numeric name or their geometric location. Particularly in graphic user interfaces, it is common to identify objects geometrically by selection with the mouse. As the number of geometric objects in a graphic user interface grows, it becomes increasingly more important to treat the basic operations add, search and remove for geometric objects with great efficiency. Guttmann has proposed the Region-Tree (R-tree) for geometric identification in an environment which uses pages on disc as data structure. Minimal bounding rectangles are used to structure the data in such a way that neighborhood relations can be described effectively. The literature shows that the parameters which influence the efficiency of the R-trees have been studied extensively, but without conclusive results. The goal of the research which is reported in this paper is to determine reliably the parameters which significantly influence the efficiency of R-trees for geometric identification in technical drawings. In order to make this investigation conclusive, it must be performed with the best available software technology. Therefore an object-oriented software for the method is developed. This implementation is tested with technical drawings containing many thousands of geometric objects. These drawings are created automatically by a stochastic generator which is incorporated into a test bed consisting of an editor and a visualisor. This test bed is used to obtain statistics for the main factors which affect the efficiency of R-trees. The investigation shows that the following main factors which affect the efficiency can be identified reliably : number of geometric objects on the drawing the minimum und maximum number of children of a node of the tree the maximum width and height of the minimal bounding rectangles of the geometric objects relative to the size of the drawing.
In earlier research, generalized multidimensional Hilbert transforms have been constructed in m-dimensional Euclidean space, in the framework of Clifford analysis. Clifford analysis, centred around the notion of monogenic functions, may be regarded as a direct and elegant generalization to higher dimension of the theory of the holomorphic functions in the complex plane. The considered Hilbert transforms, usually obtained as a part of the boundary value of an associated Cauchy transform in m+1 dimensions, might be characterized as isotropic, since the metric in the underlying space is the standard Euclidean one. In this paper we adopt the idea of a so-called anisotropic Clifford setting, which leads to the introduction of a metric dependent m-dimensional Hilbert transform, showing, at least formally, the same properties as the isotropic one. The Hilbert transform being an important tool in signal analysis, this metric dependent setting has the advantage of allowing the adjustment of the co-ordinate system to possible preferential directions in the signals to be analyzed. A striking result to be mentioned is that the associated anisotropic (m+1)-dimensional Cauchy transform is no longer uniquely determined, but may stem from a diversity of (m+1)-dimensional "mother" metrics.
The one-dimensional continuous wavelet transform is a successful tool for signal and image analysis, with applications in physics and engineering. Clifford analysis offers an appropriate framework for taking wavelets to higher dimension. In the usual orthogonal case Clifford analysis focusses on monogenic functions, i.e. null solutions of the rotation invariant vector valued Dirac operator ∂, defined in terms of an orthogonal basis for the quadratic space Rm underlying the construction of the Clifford algebra R0,m. An intrinsic feature of this function theory is that it encompasses all dimensions at once, as opposed to a tensorial approach with products of one-dimensional phenomena. This has allowed for a very specific construction of higher dimensional wavelets and the development of the corresponding theory, based on generalizations of classical orthogonal polynomials on the real line, such as the radial Clifford-Hermite polynomials introduced by Sommen. In this paper, we pass to the Hermitian Clifford setting, i.e. we let the same set of generators produce the complex Clifford algebra C2n (with even dimension), which we equip with a Hermitian conjugation and a Hermitian inner product. Hermitian Clifford analysis then focusses on the null solutions of two mutually conjugate Hermitian Dirac operators which are invariant under the action of the unitary group. In this setting we construct new Clifford-Hermite polynomials, starting in a natural way from a Rodrigues formula which now involves both Dirac operators mentioned. Due to the specific features of the Hermitian setting, four different types of polynomials are obtained, two types of even degree and two types of odd degree. These polynomials are used to introduce a new continuous wavelet transform, after thorough investigation of all necessary properties of the involved polynomials, the mother wavelet and the associated family of wavelet kernels.
Euclidean Clifford analysis is a higher dimensional function theory offering a refinement of classical harmonic analysis. The theory is centered around the concept of monogenic functions, i.e. null solutions of a first order vector valued rotation invariant differential operator called the Dirac operator, which factorizes the Laplacian. More recently, Hermitean Clifford analysis has emerged as a new and successful branch of Clifford analysis, offering yet a refinement of the Euclidean case; it focusses on the simultaneous null solutions, called Hermitean (or h-) monogenic functions, of two Hermitean Dirac operators which are invariant under the action of the unitary group. In Euclidean Clifford analysis, the Clifford-Cauchy integral formula has proven to be a corner stone of the function theory, as is the case for the traditional Cauchy formula for holomorphic functions in the complex plane. Previously, a Hermitean Clifford-Cauchy integral formula has been established by means of a matrix approach. This formula reduces to the traditional Martinelli-Bochner formula for holomorphic functions of several complex variables when taking functions with values in an appropriate part of complex spinor space. This means that the theory of Hermitean monogenic functions should encompass also other results of several variable complex analysis as special cases. At present we will elaborate further on the obtained results and refine them, considering fundamental solutions, Borel-Pompeiu representations and the Teoderescu inversion, each of them being developed at different levels, including the global level, handling vector variables, vector differential operators and the Clifford geometric product as well as the blade level were variables and differential operators act by means of the dot and wedge products. A rich world of results reveals itself, indeed including well-known formulae from the theory of several complex variables.
Image processing has been much inspired by the human vision, in particular with regard to early vision. The latter refers to the earliest stage of visual processing responsible for the measurement of local structures such as points, lines, edges and textures in order to facilitate subsequent interpretation of these structures in higher stages (known as high level vision) of the human visual system. This low level visual computation is carried out by cells of the primary visual cortex. The receptive field profiles of these cells can be interpreted as the impulse responses of the cells, which are then considered as filters. According to the Gaussian derivative theory, the receptive field profiles of the human visual system can be approximated quite well by derivatives of Gaussians. Two mathematical models suggested for these receptive field profiles are on the one hand the Gabor model and on the other hand the Hermite model which is based on analysis filters of the Hermite transform. The Hermite filters are derivatives of Gaussians, while Gabor filters, which are defined as harmonic modulations of Gaussians, provide a good approximation to these derivatives. It is important to note that, even if the Gabor model is more widely used than the Hermite model, the latter offers some advantages like being an orthogonal basis and having better match to experimental physiological data. In our earlier research both filter models, Gabor and Hermite, have been developed in the framework of Clifford analysis. Clifford analysis offers a direct, elegant and powerful generalization to higher dimension of the theory of holomorphic functions in the complex plane. In this paper we expose the construction of the Hermite and Gabor filters, both in the classical and in the Clifford analysis framework. We also generalize the concept of complex Gaussian derivative filters to the Clifford analysis setting. Moreover, we present further properties of the Clifford-Gabor filters, such as their relationship with other types of Gabor filters and their localization in the spatial and in the frequency domain formalized by the uncertainty principle.
This contribution will be freewheeling in the domain of signal, image and surface processing and touch briefly upon some topics that have been close to the heart of people in our research group. A lot of the research of the last 20 years in this domain that has been carried out world wide is dealing with multiresolution. Multiresolution allows to represent a function (in the broadest sense) at different levels of detail. This was not only applied in signals and images but also when solving all kinds of complex numerical problems. Since wavelets came into play in the 1980's, this idea was applied and generalized by many researchers. Therefore we use this as the central idea throughout this text. Wavelets, subdivision and hierarchical bases are the appropriate tools to obtain these multiresolution effects. We shall introduce some of the concepts in a rather informal way and show that the same concepts will work in one, two and three dimensions. The applications in the three cases are however quite different, and thus one wants to achieve very different goals when dealing with signals, images or surfaces. Because completeness in our treatment is impossible, we have chosen to describe two case studies after introducing some concepts in signal processing. These case studies are still the subject of current research. The first one attempts to solve a problem in image processing: how to approximate an edge in an image efficiently by subdivision. The method is based on normal offsets. The second case is the use of Powell-Sabin splines to give a smooth multiresolution representation of a surface. In this context we also illustrate the general method of construction of a spline wavelet basis using a lifting scheme.
A UNIFIED APPROACH FOR THE TREATMENT OF SOME HIGHER DIMENSIONAL DIRAC TYPE EQUATIONS ON SPHERES
(2010)
Using Clifford analysis methods, we provide a unified approach to obtain explicit solutions of some partial differential equations combining the n-dimensional Dirac and Euler operators, including generalizations of the classical time-harmonic Maxwell equations. The obtained regular solutions show strong connections between hypergeometric functions and homogeneous polynomials in the kernel of the Dirac operator.
In this paper we study the structure of the solutions to higher dimensional Dirac type equations generalizing the known λ-hyperholomorphic functions, where λ is a complex parameter. The structure of the solutions to the system of partial differential equations (D- λ) f=0 show a close connection with Bessel functions of first kind with complex argument. The more general system of partial differential equations that is considered in this paper combines Dirac and Euler operators and emphasizes the role of the Bessel functions. However, contrary to the simplest case, one gets now Bessel functions of any arbitrary complex order.
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.
Design activity could be treated as state transition computationally. In stepwise processing, in-between form-states are not easily observed. However, in this research time-based concept is introduced and applied in order to bridge the gap. In architecture, folding is one method of form manipulation and architects also want to search for alternatives by this operation. Besides, folding operation has to be defined and parameterized before time factor is involved as a variable of folding. As a result, time-based transformation provides sequential form states and redirects design activity.
MICROPLANE MODEL WITH INITIAL AND DAMAGE-INDUCED ANISOTROPY APPLIED TO TEXTILE-REINFORCED CONCRETE
(2010)
The presented material model reproduces the anisotropic characteristics of textile reinforced concrete in a smeared manner. This includes both the initial anisotropy introduced by the textile reinforcement, as well as the anisotropic damage evolution reflecting fine patterns of crack bridges. The model is based on the microplane approach. The direction-dependent representation of the material structure into oriented microplanes provides a flexible way to introduce the initial anisotropy. The microplanes oriented in a yarn direction are associated with modified damage laws that reflect the tension-stiffening effect due to the multiple cracking of the matrix along the yarn.
In this paper we consider the time independent Klein-Gordon equation on some conformally flat 3-tori with given boundary data. We set up an explicit formula for the fundamental solution. We show that we can represent any solution to the homogeneous Klein-Gordon equation on the torus as finite sum over generalized 3-fold periodic elliptic functions that are in the kernel of the Klein-Gordon operator. Furthermore we prove Cauchy and Green type integral formulas and set up a Teodorescu and Cauchy transform for the toroidal Klein-Gordon operator. These in turn are used to set up explicit formulas for the solution to the inhomogeneous version of the Klein-Gordon equation on the 3-torus.
ON THE NAVIER-STOKES EQUATION WITH FREE CONVECTION IN STRIP DOMAINS AND 3D TRIANGULAR CHANNELS
(2006)
The Navier-Stokes equations and related ones can be treated very elegantly with the quaternionic operator calculus developed in a series of works by K. Guerlebeck, W. Sproeossig and others. This study will be extended in this paper. In order to apply the quaternionic operator calculus to solve these types of boundary value problems fully explicitly, one basically needs to evaluate two types of integral operators: the Teodorescu operator and the quaternionic Bergman projector. While the integral kernel of the Teodorescu transform is universal for all domains, the kernel function of the Bergman projector, called the Bergman kernel, depends on the geometry of the domain. With special variants of quaternionic holomorphic multiperiodic functions we obtain explicit formulas for three dimensional parallel plate channels, rectangular block domains and regular triangular channels. The explicit knowledge of the integral kernels makes it then possible to evaluate the operator equations in order to determine the solutions of the boundary value problem explicitly.
In classical complex function theory the geometric mapping property of conformality is closely linked with complex differentiability. In contrast to the planar case, in higher dimensions the set of conformal mappings is only the set of Möbius transformations. Unfortunately, the theory of generalized holomorphic functions (by historical reasons they are called monogenic functions) developed on the basis of Clifford algebras does not cover the set of Möbius transformations in higher dimensions, since Möbius transformations are not monogenic. But on the other side, monogenic functions are hypercomplex differentiable functions and the question arises if from this point of view they can still play a special role for other types of 3D-mappings, for instance, for quasi-conformal ones. On the occasion of the 16th IKM 3D-mapping methods based on the application of Bergman's reproducing kernel approach (BKM) have been discussed. Almost all authors working before that with BKM in the Clifford setting were only concerned with the general algebraic and functional analytic background which allows the explicit determination of the kernel in special situations. The main goal of the abovementioned contribution was the numerical experiment by using a Maple software specially developed for that purpose. Since BKM is only one of a great variety of concrete numerical methods developed for mapping problems, our goal is to present a complete different from BKM approach to 3D-mappings. In fact, it is an extension of ideas of L. V. Kantorovich to the 3-dimensional case by using reduced quaternions and some suitable series of powers of a small parameter. Whereas until now in the Clifford case of BKM the recovering of the mapping function itself and its relation to the monogenic kernel function is still an open problem, this approach avoids such difficulties and leads to an approximation by monogenic polynomials depending on that small parameter.
In the past, several types of Fourier transforms in Clifford analysis have been studied. In this paper, first an overview of these different transforms is given. Next, a new equation in a Clifford algebra is proposed, the solutions of which will act as kernels of a new class of generalized Fourier transforms. Two solutions of this equation are studied in more detail, namely a vector-valued solution and a bivector-valued solution, as well as the associated integral transforms.
THE FOURIER-BESSEL TRANSFORM
(2010)
In this paper we devise a new multi-dimensional integral transform within the Clifford analysis setting, the so-called Fourier-Bessel transform. It appears that in the two-dimensional case, it coincides with the Clifford-Fourier and cylindrical Fourier transforms introduced earlier. We show that this new integral transform satisfies operational formulae which are similar to those of the classical tensorial Fourier transform. Moreover the L2-basis elements consisting of generalized Clifford-Hermite functions appear to be eigenfunctions of the Fourier-Bessel transform.
Designing a structure follows a pattern of creating a structural design concept, executing a finite element analysis and developing a design model. A project was undertaken to create computer support for executing these tasks within a collaborative environment. This study focuses on developing a software architecture that integrates the various structural design aspects into a seamless functional collaboratory that satisfies engineering practice requirements. The collaboratory is to support both homogeneous collaboration i.e. between users operating on the same model and heterogeneous collaboration i.e. between users operating on different model types. Collaboration can take place synchronously or asynchronously, and the information exchange is done either at the granularity of objects or at the granularity of models. The objective is to determine from practicing engineers which configurations they regard as best and what features are essential for working in a collaborative environment. Based on the suggestions of these engineers a specification of a collaboration configuration that satisfies engineering practice requirements will be developed.
This paper describes the application of interval calculus to calculation of plate deflection, taking in account inevitable and acceptable tolerance of input data (input parameters). The simply supported reinforced concrete plate was taken as an example. The plate was loaded by uniformly distributed loads. Several parameters that influence the plate deflection are given as certain closed intervals. Accordingly, the results are obtained as intervals so it was possible to follow the direct influence of a change of one or more input parameters on output (in our example, deflection) values by using one model and one computing procedure. The described procedure could be applied to any FEM calculation in order to keep calculation tolerances, ISO-tolerances, and production tolerances in close limits (admissible limits). The Wolfram Mathematica has been used as tool for interval calculation.
Reasonably accurate cost estimation of the structural system is quite desirable at the early stages of the design process of a construction project. However, the numerous interactions among the many cost-variables make the prediction difficult. Artificial neural networks (ANN) and case-based reasoning (CBR) are reported to overcome this difficulty. This paper presents a comparison of CBR and ANN augmented by genetic algorithms (GA) conducted by using spreadsheet simulations. GA was used to determine the optimum weights for the ANN and CBR models. The cost data of twenty-nine actual cases of residential building projects were used as an example application. Two different sets of cases were randomly selected from the data set for training and testing purposes. Prediction rates of 84% in the GA/CBR study and 89% in the GA/ANN study were obtained. The advantages and disadvantages of the two approaches are discussed in the light of the experiments and the findings. It appears that GA/ANN is a more suitable model for this example of cost estimation where the prediction of numerical values is required and only a limited number of cases exist. The integration of GA into CBR and ANN in a spreadsheet format is likely to improve the prediction rates.
Der Begriff der Zuverlässigkeit spielt eine zentrale Rolle bei der Bewertung von Verkehrsnetzen. Aus der Sicht der Nutzer des öffentlichen Personennahverkehrs (ÖPNV) ist eines der wichtigsten Kriterien zur Beurteilung der Qualität des Liniennetzes, ob es möglich ist, mit einer großen Sicherheit das Reiseziel in einer vorgegebenen Zeit zu erreichen. Im Vortrag soll dieser Zuverlässigkeitsbegriff mathematisch gefasst werden. Dabei wird zunächst auf den üblichen Begriff der Zuverlässigkeit eines Netzes im Sinne paarweiser Zusammenhangswahrscheinlichkeiten eingegangen. Dieser Begriff wird erweitert durch die Betrachtung der Zuverlässigkeit unter Einbeziehung einer maximal zulässigen Reisezeit. In vergangenen Arbeiten hat sich die Ring-Radius-Struktur als bewährtes Modell für die theoretische Beschreibung von Verkehrsnetzen erwiesen. Diese Überlegungen sollen nun durch Einbeziehung realer Verkehrsnetzstrukturen erweitert werden. Als konkretes Beispiel dient das Straßenbahnnetz von Krakau. Hier soll insbesondere untersucht werden, welche Auswirkungen ein geplanter Ausbau des Netzes auf die Zuverlässigkeit haben wird. This paper is involved with CIVITAS-CARAVEL project: "Clean and better transport in cites". The project has received research funding from the Community's Sixth Framework Programme. The paper reflects only the author's views and the Community is not liable for any use that may be made of the information contained therein.
Wir betrachten im ÖPNV (Öffentlichen Personennahverkehr) diejenige Situation, daß zwei Bus- oder Straßenbahnlinien gemeinsame Haltestellen haben. Ziel unserer Untersuchungen ist es, für beide Linien einen solchen Fahrplan zu finden, der für die Fahrgäste möglichst viel Bequemlichkeit bietet. Die Bedarfsstruktur - die Anzahl von Personen, die die beiden Linien benutzen - setzt dabei gewisse Beschränkungen für die Taktzeiten der beiden Linien. Die verbleibenden Entscheidungsfreiheiten sollen im Sinne der Zielstellung ausgenutzt werden. Im Vortrag wird folgenden Fragen nachgegangen: - nach welchen Kriterien kann man die "Bequemlichkeit" oder die "Synchonisationsgüte" messen? - wie kann man die einzelnen "Synchronisationsmaße" berechnen ? - wie kann man die verbleibenden Entscheidungsfreiheiten nutzen, um eine möglichst gute Synchronisation zu erreichen ? Die Ergebnisse werden dann auf einige Beispiele angewandt und mit den bereitgestellten Methoden Lösungsvorschläge unterbreitet.
In civil engineering it is very difficult and often expensive to excite constructions such as bridges and buildings with an impulse hammer or shaker. This problem can be avoided with the output-only method as special feature of stochastic system identification. The permanently existing ambient noise (e.g. wind, traffic, waves) is sufficient to excite the structures in their operational conditions. The output-only method is able to estimate the observable part of a state-space-model which contains the dynamic characteristics of the measured mechanical system. Because of the assumption that the ambient excitation is white there is no requirement to measure the input. Another advantage of the output-only method is the possibility to get high detailed models by a special method, called polyreference setup. To pretend the availability of a much larger set of sensors the data from varying sensor locations will be collected. Several successive data sets are recorded with sensors at different locations (moving sensors) and fixed locations (reference sensors). The covariance functions of the reference sensors are bases to normalize the moving sensors. The result of the following subspace-based system identification is a high detailed black-box-model that contains the weighting function including the well-known dynamic parameters eigenfrequencies and mode shapes of the mechanical system. Emphasis of this lecture is the presentation of an extensive damage detection experiment. A 53-year old prestressed concrete tied-arch-bridge in Hünxe (Germany) was deconstructed in 2005. Preliminary numerous vibration measurements were accomplished. The first experiment for system modification was an additional support near the bridge bearing of one main girder. During a further experiment one hanger from one tied arch was cut through as an induced damage. Some first outcomes of the described experiments will be presented.
NONZONAL WAVELETS ON S^N
(2010)
In the present article we will construct wavelets on an arbitrary dimensional sphere S^n due the approach of approximate Identities. There are two equivalently approaches to wavelets. The group theoretical approach formulates a square integrability condition for a group acting via unitary, irreducible representation on the sphere. The connection to the group theoretical approach will be sketched. The concept of approximate identities uses the same constructions in the background, here we select an appropriate section of dilations and translations in the group acting on the sphere in two steps. At First we will formulate dilations in terms of approximate identities and than we call in translations on the sphere as rotations. This leads to the construction of an orthogonal polynomial system in L²(SO(n+1)). That approach is convenient to construct concrete wavelets, since the appropriate kernels can be constructed form the heat kernel leading to the approximate Identity of Gauss-Weierstra\ss. We will work out conditions to functions forming a family of wavelets, subsequently we formulate how we can construct zonal wavelets from a approximate Identity and the relation to admissibility of nonzonal wavelets. Eventually we will give an example of a nonzonal Wavelet on $S^n$, which we obtain from the approximate identity of Gauss-Weierstraß.
Due to the amount of flow simulation and measurement data, automatic detection, classification and visualization of features is necessary for an inspection. Therefore, many automated feature detection methods have been developed in recent years. However, only one feature class is visualized afterwards in most cases, and many algorithms have problems in the presence of noise or superposition effects. In contrast, image processing and computer vision have robust methods for feature extraction and computation of derivatives of scalar fields. Furthermore, interpolation and other filter can be analyzed in detail. An application of these methods to vector fields would provide a solid theoretical basis for feature extraction. The authors suggest Clifford algebra as a mathematical framework for this task. Clifford algebra provides a unified notation for scalars and vectors as well as a multiplication of all basis elements. The Clifford product of two vectors provides the complete geometric information of the relative positions of these vectors. Integration of this product results in Clifford correlation and convolution which can be used for template matching of vector fields. For frequency analysis of vector fields and the behavior of vector-valued filters, a Clifford Fourier transform has been derived for 2D and 3D. Convolution and other theorems have been proved, and fast algorithms for the computation of the Clifford Fourier transform exist. Therefore the computation of Clifford convolution can be accelerated by computing it in Clifford Fourier domain. Clifford convolution and Fourier transform can be used for a thorough analysis and subsequent visualization of flow fields.
Die Liquiditätsplanung von Bauunternehmen XE "Liquiditätsplanung" gilt als ein wesentliches Steuerungs-, Kontroll- sowie Informationsinstrument für interne und externe Adressaten und übt eine Entscheidungsunterstützungsfunktion aus. Da die einzelnen Bauprojekte einen wesentlichen Anteil an den Gesamtkosten des Unternehmens ausmachen, besitzen diese auch einen erheblichen Einfluß auf die Liquidität und die Zahlungsfähigkeit der Bauunternehmung. Dem folgend ist es in der Baupraxis eine übliche Verfahrensweise, die Liquiditätsplanung zuerst projektbezogen zu erstellen und anschließend auf Unternehmensebene zu verdichten. Ziel der Ausführungen ist es, die Zusammenhänge von Arbeitskalkulation XE "Arbeitskalkulation" , Ergebnisrechnung XE "Ergebnisrechnung" und Finanzrechnung XE "Finanzrechnung" in Form eines deterministischen XE "Erklärungsmodells" Planungsmodells auf Projektebene darzustellen. Hierbei soll das Verständnis und die Bedeutung der Verknüpfungen zwischen dem technisch-orientierten Bauablauf und dessen Darstellung im Rechnungs- und Finanzwesen herausgestellt werden. Die Vorgänge aus der Bauabwicklung, das heißt die Abarbeitung der Bauleistungsverzeichnispositionen und deren zeitliche Darstellung in einem Bauzeitenplan sind periodisiert in Größen der Betriebsbuchhaltung (Leistung, Kosten) zu transformieren und anschließend in der Finanzrechnung (Einzahlungen., Auszahlungen) nach Kreditoren und Debitoren aufzuschlüsseln.
We consider efficient numerical methods for the solution of partial differential equations with stochastic coefficients or right hand side. The discretization is performed by the stochastic finite element method (SFEM). Separation of spatial and stochastic variables in the random input data is achieved via a Karhunen-Loève expansion or Wiener's polynomial chaos expansion. We discuss solution strategies for the Galerkin system that take advantage of the special structure of the system matrix. For stochastic coefficients linear in a set of independent random variables we employ Krylov subspace recycling techniques after having decoupled the large SFEM stiffness matrix.
In many branches companies often lose the visibility of their human and technical resources of their field service. On the one hand the people in the fieldservice are often free like kings on the other hand they do not take part of the daily communication in the central office and suffer under the lacking involvement in the decisions inside the central office. The result is inefficiency. Reproaches in both directions follow. With the radio systems and then mobile phones the ditch began to dry up. But the solutions are far from being productive.
HYPERMONOGENIC POLYNOMIALS
(2006)
It is well know that the power function is not monogenic. There are basically two ways to include the power function into the set of solutions: The hypermonogenic functions or holomorphic Cliffordian functions. L. Pernas has found out the dimension of the space of homogenous holomorphic Cliffordian polynomials of degree m, but his approach did not include a basis. It is known that the hypermonogenic functions are included in the space of holomorphic Cliffordian functions. As our main result we show that we can construct a basis for the right module of homogeneous holomorphic Cliffordian polynomials of degree m using hypermonogenic polynomials and their derivatives. To that end we first recall the function spaces of monogenic, hypermonogenic and holomorphic Cliffordian functions and give the results needed in the proof of our main theorem. We list some basic polynomials and their properties for the various function spaces. In particular, we consider recursive formulas, rules of differentiation and properties of linear independency for the polynomials.
Traffic simulation is a valuable tool for the design and evaluation of road networks. Over the years, the level of detail to which urban and freeway traffic can be simulated has increased steadily, shifting from a merely qualitative macroscopic perspective to a very detailed microscopic view, where the behavior of individual vehicles is emulated realistically. With the improvement of behavioral models, however, the computational complexity has also steadily increased, as more and more aspects of real-life traffic have to be considered by the simulation environment. Despite the constant increase in computing power of modern personal computers, microscopic simulation stays computationally expensive, limiting the maximum network size than can be simulated on a single-processor computer in reasonable time. Parallelization can distribute the computing load from a single computer system to a cluster of several computing nodes. To this end, the exisiting simulation framework had to be adapted to allow for a distributed approach. As the simulation is ultimately targeted to be executed in real-time, incorporating real traffic data, only a spatial partition of the simulation was considered, meaning the road network has to be partitioned into subnets of comparable complexity, to ensure a homogenous load balancing. The partition process must also ensure, that the division between subnets does only occur in regions, where no strong interaction between the separated road segments occurs (i.e. not in the direct vicinity of junctions). In this paper, we describe a new microscopic reasoning voting strategy, and discuss in how far the increasing computational costs of these more complex behaviors lend themselves to a parallelized approach. We show the parallel architecture employed, the communication between computing units using MPIJava, and the benefits and pitfalls of adapting a single computer application to be used on a multi-node computing cluster.
The use of process models in the analysis, optimization and simulation of processes has proven to be extremely beneficial in the instances where they could be applied appropriately. However, the Architecture/Engineering/Construction (AEC) industries present unique challenges that complicate the modeling of their processes. A simple Engineering process model, based on the specification of Tasks, Datasets, Persons and Tools, and certain relations between them, have been developed, and its advantages over conventional techniques have been illustrated. Graph theory is used as the mathematical foundation mapping Tasks, Datasets, Persons and Tools to vertices and the relations between them to edges forming a directed graph. The acceptance of process modeling in AEC industries not only depends on the results it can provide, but the ease at which these results can be attained. Specifying a complex AEC process model is a dynamic exercise that is characterized by many modifications over the process model's lifespan. This article looks at reducing specification complexity, reducing the probability for erroneous input and allowing consistent model modification. Furthermore, the problem of resource leveling is discussed. Engineering projects are often executed with limited resources and determining the impact of such restrictions on the sequence of Tasks is important. Resource Leveling concerns itself with these restrictions caused by limited resources. This article looks at using Task shifting strategies to find a near-optimal sequence of Tasks that guarantees consistent Dataset evolution while resolving resource restrictions.
In this paper we consider three different methods for generating monogenic functions. The first one is related to Fueter's well known approach to the generation of monogenic quaternion-valued functions by means of holomorphic functions, the second one is based on the solution of hypercomplex differential equations and finally the third one is a direct series approach, based on the use of special homogeneous polynomials. We illustrate the theory by generating three different exponential functions and discuss some of their properties. Formula que se usa em preprints e artigos da nossa UI&D (acho demasiado completo): Partially supported by the R\&D unit \emph{Matem\'atica a Aplica\c\~es} (UIMA) of the University of Aveiro, through the Portuguese Foundation for Science and Technology (FCT), co-financed by the European Community fund FEDER.
We establish the basis of a discrete function theory starting with a Fischer decomposition for difference Dirac operators. Discrete versions of homogeneous polynomials, Euler and Gamma operators are obtained. As a consequence we obtain a Fischer decomposition for the discrete Laplacian. For the sake of simplicity we consider in the first part only Dirac operators which contain only forward or backward finite differences. Of course, these Dirac operators do not factorize the classic discrete Laplacian. Therefore, we will consider a different definition of a difference Dirac operator in the quaternionic case which do factorizes the discrete Laplacian.
SIMULATION AND MATHEMATICAL OPTIMIZATION OF THE HYDRATION OF CONCRETE FOR AVOIDING THERMAL CRACKS
(2010)
After mixing of concrete, the hardening starts by an exothermic chemical reaction known as hydration. As the reaction rate depends on the temperature the time in the description of the hydration is replaced by the maturity which is defined as an integral over a certain function depending on the temperature. The temperature distribution is governed by the heat equation with a right hand side depending on the maturity and the temperature itself. We compare of the performance of different time integration schemes of higher order with an automatic time step control. The simulation of the heat distribution is of importance as the development of mechanical properties is driven by the hydration. During this process it is possible that the tensile stresses exceed the tensile strength and cracks occur. The goal is to produce cheap concrete without cracks. Simple crack-criterions use only temperature differences, more involved ones are based on thermal stresses. If the criterion predicts cracks some changes in the input data are needed. This can be interpreted as optimization. The final goal will be to adopt model based optimization (in contrast to simulation based optimization) to the problem of the hydration of young concrete and the avoidance of cracks. The first step is the simulation of the hydration, which we focus in this paper.
An introduction is given to Clifford Analysis over pseudo-Euclidean space of arbitrary signature, called for short Ultrahyperbolic Clifford Analysis (UCA). UCA is regarded as a function theory of Clifford-valued functions, satisfying a first order partial differential equation involving a vector-valued differential operator, called a Dirac operator. The formulation of UCA presented here pays special attention to its geometrical setting. This permits to identify tensors which qualify as geometrically invariant Dirac operators and to take a position on the naturalness of contravariant and covariant versions of such a theory. In addition, a formal method is described to construct the general solution to the aforementioned equation in the context of covariant UCA.
LIFETIME-ORIENTED OPTIMIZATION OF BRIDGE TIE RODS EXPOSED TO VORTEX-INDUCED ACROSS-WIND VIBRATIONS
(2006)
In recent years, damages in welded connections plates of vertical tie rods of several arched steel bridges have been reported. These damages are due to fatigue caused by wind-induced vibrations. In the present study, such phenomena are examined, and the corresponding lifetime of a reference bridge in Münster-Hiltrup, Germany, is estimated, based on the actual shape of the connection plate. Also, the results obtained are compared to the expected lifetime of a connection plate, whose geometry has been optimized separately. The structural optimization, focussing on the shape of the cut at the hanger ends, has been carried out using evolution strategies. The oscillation amplitudes have been computed by means of the Newmark-Wilson time-step method, using an appropriate load model, which has been validated by on-site experiments on the selected reference bridge. Corresponding stress-amplitudes are evaluated by multiplying the oscillation amplitudes with a stress concentration factor. This factor has been computed on the basis of a finite element model of the system "hanger-welding-connection plate", applying solid elements, according to the notch stress approach. The damage estimation takes into account the stochastics of the exciting wind process, as well as the stochastics of the material parameters (fatigue strength) given in terms of Woehler-curves. The shape optimization results in a substantial increase of the estimated hanger lifetime. The comparison of the lifetimes of the bulk plate and of the welding revealed that, in the optimized structure, the welding, being the most sensitive part in the original structure, shows much more resistance against potential damages than the bulk material.
The mathematical and technical foundations of optimization have been developed to a large extent. In the design of buildings, however, optimization is rarely applied because of insufficient adaptation of this method to the needs of building design. The use of design optimization requires the consideration of all relevant objectives in an interactive and multidisciplinary process. Disciplines such as structural, light, and thermal engineering, architecture, and economics impose various objectives on the design. A good solution calls for a compromise between these often contradictory objectives. This presentation outlines a method for the application of Multidisciplinary Design Optimization (MDO) as a tool for the designing of buildings. An optimization model is established considering the fact that in building design the non-numerical aspects are of major importance than in other engineering disciplines. A component-based decomposition enables the designer to manage the non-numerical aspects in an interactive design optimization process. A façade example demonstrates a way how the different disciplines interact and how the components integrate the disciplines in one optimization model. In this grid-based façade example, the materials switch between a discrete number of materials and construction types. For light and thermal engineering, architecture, and economics, analysis functions calculate the performance; utility functions serve as an important means for the evaluation since not every increase or decrease of a physical value improves the design. For experimental purposes, a genetic algorithm applied to the exemplary model demonstrates the use of optimization in this design case. A component-based representation first serves to manage non-numerical characteristics such as aesthetics. Furthermore, it complies with usual fabrication methods in building design and with object-oriented data handling in CAD. Therefore, components provide an important basis for an interactive MDO process in building design.
Reducing energy consumption is one of the major challenges for present day and will continue for future generations. The emerging EU directives relating to energy (EU EPBD and the EU Directive on Emissions Trading) now place demands on building owners to rate the energy performance of their buildings for efficient energy management. Moreover European Legislation (Directive 2006/32/EC) requires Facility Managers to reduce building energy consumption and operational costs. Currently sophisticated building services systems are available integrating off-the-shelf building management components. However this ad-hoc combination presents many difficulties to building owners in the management and upgrade of these systems. This paper addresses the need for integration concepts, holistic monitoring and analysis methodologies, life-cycle oriented decision support and sophisticated control strategies through the seamless integration of people, ICT-devices and computational resources via introducing the newly developed integrated system architecture. The first concept was applied to a residential building and the results were elaborated to improve current building conditions.
In this paper we present rudiments of a higher dimensional analogue of the Szegö kernel method to compute 3D mappings from elementary domains onto the unit sphere. This is a formal construction which provides us with a good substitution of the classical conformal Riemann mapping. We give explicit numerical examples and discuss a comparison of the results with those obtained alternatively by the Bergman kernel method.
We propose a new approach to the numerical solution of quasi-static elastic-plastic problems based on the Moreau-Yosida theorem. After the time discretization, the problem is expressed as an energy minimization problem for unknown displacement and plastic strain fields. The dependency of the minimization functional on the displacement is smooth whereas the dependency on the plastic strain is non-smooth. Besides, there exists an explicit formula, how to calculate the plastic strain from a given displacement field. This allows us to reformulate the original problem as a minimization problem in the displacement only. Using the Moreau-Yosida theorem from the convex analysis, the minimization functional in the displacements turns out to be Frechet-differentiable, although the hidden dependency on the plastic strain is non-differentiable. The seconds derivative exists everywhere apart from the elastic-plastic interface dividing elastic and plastic zones of the continuum. This motivates to implement a Newton-like method, which converges super-linearly as can be observed in our numerical experiments.
RESEARCH OF DEFORMATION OF MULTILAYERED PLATES ON UNDEFORMABLE BASIS BY UNFLEXURAL SPECIFIED MODEL
(2006)
Stress-strain state (SSS) of multilayered plates on undeformable foundation is investigated. The settlement circuit of transverse loaded plate is formed by symmetrical attaching of a plate concerning a surface of contact to the foundation. The plate of the double thickness becomes bilateral symmetrically loaded concerning its median surface. It allows to model only unflexural deformation that reduces amount of unknown and the general order of differentiation of resolving system of the equations. The developed refined continual model takes into account deformations of transverse shear and transverse compression in high iterative approximation. Rigid contact between the foundation and a plate, and also shear without friction on a surface of contact of a plate with the foundation is considered. Calculations confirm efficiency of such approach, allowing to receive decisions which is qualitative and quantitatively close to three-dimensional solutions.
Using a quaternionic reformulation of the electrical impedance equation, we consider a two-dimensional separable-variables conductivity function and, posing two different techniques, we obtain a special class of Vekua equation, whose general solution can be approach by virtue of Taylor series in formal powers, for which is possible to introduce an explicit Bers generating sequence.
Solid behavior as well as liquid behavior characterizes the flow of granular material in silos. The presented model is based on an appropriate interaction of a displacement field and a velocity field. The constitutive equations and the applied algorithm are developed from the exact solution for a standard case. The standard case evolves from a very tall vertical plane strain silo containing material that flows at a constant speed. No horizontal displacements and velocities take place. No changes regarding the field values arise in the vertical direction and in time. Tension is not allowed at any point. Coulomb friction represents the effects of the vertical walls. The interaction between the flowing material and the walls is covered by a forced boundary condition resulting in an additional matrix for the solid component as well as for the liquid component. The resulting integral equations are designed to be solved directly. Three coefficients describe the properties of the granular material. They govern elastic solid behavior in combination with viscous liquid behavior.
Adopting the European laws concerning environmental protection will require sustained efforts of the authorities and communities from Romania; implementing modern solutions will become a fast and effective option for the improvement of the functioning systems, in order to prevent disasters. As a part of the urban infrastructure, the drainage networks of pluvial and residual waters are included in the plan of promoting the systems which protect the environmental quality, with the purpose of integrated and adaptive management. The paper presents a distributed control system for sewer network of Iasi town. Unsatisfactory technical state of the actual sewer system is exposed, focusing on objectives related to implementation of the control system. The proposed distributed control system of Iasi drainage network is based on the implementation of the hierarchic control theory for diagnose, sewer planning and management. There are proposed two control levels: coordinating and local execution. Configuration of the distributed control system, including data acquisition and conversion equipment, interface characteristics, local data bus, data communication network, station configuration are widely described. The project wish to be an useful instrument for the local authorities in the preventing and reducing the impact of future natural disasters over the urban areas by means of modern technologies.
In this paper three different formulations of a Bernoulli type free boundary problem are discussed. By analyzing the shape Hessian in case of matching data it is distinguished between well-posed and ill-posed formulations. A nonlinear Ritz-Galerkin method is applied for discretizing the shape optimization problem. In case of well-posedness existence and convergence of the approximate shapes is proven. In combination with a fast boundary element method efficient first and second order shape optimization algorithms are obtained.
DECENTRALIZED APPROACHES TO ADAPTIVE TRAFFIC CONTROL AND AN EXTENDED LEVEL OF SERVICE CONCEPT
(2006)
Traffic systems are highly complex multi-component systems suffering from instabilities and non-linear dynamics, including chaos. This is caused by the non-linearity of interactions, delays, and fluctuations, which can trigger phenomena such as stop-and-go waves, noise-induced breakdowns, or slower-is-faster effects. The recently upcoming information and communication technologies (ICT) promise new solutions leading from the classical, centralized control to decentralized approaches in the sense of collective (swarm) intelligence and ad hoc networks. An interesting application field is adaptive, self-organized traffic control in urban road networks. We present control principles that allow one to reach a self-organized synchronization of traffic lights. Furthermore, vehicles will become automatic traffic state detection, data management, and communication centers when forming ad hoc networks through inter-vehicle communication (IVC). We discuss the mechanisms and the efficiency of message propagation on freeways by short-range communication. Our main focus is on future adaptive cruise control systems (ACC), which will not only increase the comfort and safety of car passengers, but also enhance the stability of traffic flows and the capacity of the road (“traffic assistance”). We present an automated driving strategy that adapts the operation mode of an ACC system to the autonomously detected, local traffic situation. The impact on the traffic dynamics is investigated by means of a multi-lane microscopic traffic simulation. The simulation scenarios illustrate the efficiency of the proposed driving strategy. Already an ACC equipment level of 10% improves the traffic flow quality and reduces the travel times for the drivers drastically due to delaying or preventing a breakdown of the traffic flow. For the evaluation of the resulting traffic quality, we have recently developed an extended level of service concept (ELOS). We demonstrate our concept on the basis of travel times as the most important variable for a user-oriented quality of service.
The concrete is modeled as a material with damage and plasticity, whereat the viscoplastic and the viscoelastic behaviour depends on the rate of the total strains. Due to the damage behaviour the compliance tensor develops different properties in tension and compression. There have been tested various yield surfaces and flow rules, damage rules respectively to their usability in a concrete model. One three-dimensional yield surface was developed from a failure surface based on the Willam--Warnke five-parameter model by the author. Only one general uni-axial stress-strain-relation is used for the numeric control of the yield surface. From that curve all necessary parameters for different strengths of concrete and different strain rates can be derived by affine transformations. For the flow rule in the compression zone a non associated inelastic potential is used, in the tension zone a Rankine potential. Conditional on the time-dependent formulation, the symmetry of the system equations is maintained in spite of the usage of non-associated potentials for the derivation of the inelastic strains. In case of quasi statical computations a simple viscoplastic law is used that is rested on an approach to Perzyna. The principle of equality of dissipation power in the uni-axial and the three-axial state of stress is used. It is modified by a factor that depends on the actual stress ratio and in comparison with the Kupfer experiments it implicates strains that are more realistic. The implementation of the concrete model is conducted in a mixed hybrid finite element. Examples in the structural level are introduced for verification of the concrete model.
Digital models of buildings are widely used in civil engineering. In these models, geometric information is used as leading information. Engineers are used to have geometric information, and, for instance, it is state of the art to specify a point by its three coordinates. However, the traditional approaches have disadvantages. Geometric information is over-determined. Thus, more geometric information is specified and stored than needed. In addition, engineers already deal with topological information. A denotation of objects in buildings is of topological nature. It has to be answered whether approaches where topological information becomes a leading role would be more efficient in civil engineering. This paper presents such an approach. Topological information is modelled independently of geometric information. It is used for denoting the objects of a building. Geometric information is associated to topological information so that geometric information “weights” a topology.
The concept presented in this paper has already been used in surveying existing buildings. Experiences in the use of this concept showed that the number of geometric information that is required for a complete specification of a building could be reduced by a factor up to 100. Further research will show how this concept can be used in planning processes.
Planning and construction processes are characterized by the peculiarity that they need to be designed individually for each project. It is necessary to set up an individual schedule for each project. As a basis for a new project, schedules from already finished projects are used, but adaptions are always necessary. In practice, scheduling tools only document a process. Schedules cover a set of activities, their duration and a set of interdependencies between activities. The design of a process is up to the user. It is not necessary to specify each interdependency, and completeness and correctness need to be checked manually. No methodologies are available to guarantee properties such as correctness or completeness. The considerations presented in the paper are based on an approach where a planning and a construction process including the interdependencies between planning and construction activities are regarded as a result. Selected information need to be specified by a user, and a proposal for an order of planning and construction activities is computed. As a consequence, process properties such as correctness and completeness can be guaranteed with respect to user input. Especially in Germany, clients are allowed to modify their requirements at any time. This leads to modifications in the planning and construction processes. This paper covers a mathematical formulation for this problem based on set theory. A complex structure is set up covering objects and relations; and operations are defined that guarantee consistency in the underlying and versioned process description. The presented considerations are based on previous work. This paper can be regarded as the next step in a series of previous work describing how a suitable concept for handling, planning and construction processes in civil engineering can be formed.