SIMULATION

Author: CAMPO VAZQUEZ M. CELESTE.
Year: 2003.
University: CARLOS III DE MADRID [www.uc3m.es].
Place of defense: ESCUELA POLITÉCNICA SUPERIOR.
Place of preparation: UNIVERSIDAD CARLOS III DE MADRID.

Summary: This doctoral thesis we made contributions in the field of the definition of middleware technology for the development of services in ubiquitous computing environments. First, we address the problem of discovery service, which will enable a device to automatically discover the services offered by other devices that surround her. Although there are proposals in this regard, do not cover all requirements imposed by these new computing environments, so we have definidio a new mechanism of discovery: Pervasive Discovery Protocol (PDP). In sengudo place, we propose the use of technology such as mobile agents middleware for the development of services in ubiquitous environments. The agents are characterized as being oriented tasks, being self-employed, for their ability to cooperate and whether they have the characteristic of mobility, being able to move to remote systems to accomplish their tasks, while minimizing the cost of communications. These characteristics are adapted to the constraints imposed pervasive computing. In this line, contributed to the adaptation of standard IFAP agents to operate in these environments, and in particular, we focus on the Directory Faciliator (DF). These proposals have led us to actively participate in the ad hoc technical committee IFAP.

Author: PEGUERO CHAMIZO JUAN CARLOS.
Year: 2003.
University: EXTREMADURA [www.unex.es].
Place of defense: CENTRO UNIVERSITARIO DE MERIDA.
Place of preparation: ESCUELA DE INGENIERÍAS INDUSTRIALES.

Summary: This thesis has been the simulated biological reactor of a waste water treatment plant (WWTP). Its dynamics has been controlled by a neural network has learned that the actions of technical control that governs real plant. This network operates on the constant recirculation biolados reactor. To maximize the response has designed an iterative procedure for adjusting the constant bearing the dynamics of the plant to an area of work optimally. The adjustment values are learned by a second neural network that is able to predict the optimum dela recirculation two neural networks used are convinadas through a third designed with fixed parameters that causes the effects of the previous two.

Author: GIL SOTO JOSÉ JAVIER.
Year: 2004.
University: NAVARRA [www.unav.es].
Place of defense: ESCUELA TÉCNICA SUPERIOR DE INGENIEROS INDUSTRIALES Y DE TELECOMUNICACIÓN.
Place of preparation: UNIVERSIDAD PÚBLICA DE NAVARRA.

Summary: The work develops focuses on the simulation of mechanical systems. It consists of the conceptual development and subsequent implementation of a symbolic system that allows for simple common problems of mechanics clásica.El system is complemented by a symbolic number system that allows resolution of the various numerical simulations raised. The developed symbolic system is based on symbolic algebra system Matlab / Maple, and the numbering system in the numbering system Matlab, both of proven reliability. The "proof of concept" development is limited to the case of systems of rigid bodies, but the philosophy of the implementation makes it easily extensible systems combined rígido-flexible. Although not the primary objective, the program develops serves as a tool in teaching subjects such as the Theory of Machines, or the design and calculation of machines.

Author: POVEDA ALBALADEJO JOSE FRANCISCO.
Year: 2004.
University: JAUME I DE CASTELLON [www.uji.es].
Place of defense: VILLA ELISA BENICASIM (CASTELLON).
Place of preparation: ESCUELA SUPERIOR DE TECNOLOGIA Y CIENCIAS EXPERIMENTALES.

Summary: In this paper we address problems and proposed solutions associated with the generation and representation of Digital Terrain Models. Specifically, in relation to the generation of synthetic ground proposes a model for the design and generation of mounds beginning of a process of interpolation through the electrostatic field lines associated with the map of the terrain designed isolines. Associated with this process presents the algorithm BQV, a new fast algorithm for the relative location of a point over a polygon defined on the plane, not necessarily regulate, or related, or convex. With regard to the representation and interactive three-dimensional terrain is so pervasive in a space of arbitrary dimension, a characterization matrix to calculate the vicinity directional trees in space division. Finally, in order to optimize the delivery of database space associated with the model simplification and visualization dependent on the hearing presented deals with the design and analysis of different schemes for transforming textures. As a consequence of the results obtained is proposed for the database space two possible schemes.

Author: POVEDA ALBALADEJO JOSE FRANCISCO.
Year: 2004.
University: JAUME I DE CASTELLON [www.uji.es].
Place of defense: VILLA ELISA BENICASIM.
Place of preparation: E.S. TECNOLOGIA Y CIENCIAS EXPERIMENTALES.

Summary: In this paper we address problems and proposed solutions associated with the generation and representation of Digital Terrain Models. Specifically, in relation to the generation of synthetic ground proposes a model for the design and generation of mounds beginning of a process of interpolation through the electrostatic field lines associated with the map of the terrain designed isolines. Associated with this process presents the algorithm BQV, a new fast algorithm for the relative location of a point over a polygon defined on the plane, not necessarily regulate, or related, or convex. With regard to the representation and interactive three-dimensional terrain is so pervasive in a space of arbitrary dimension, a characterization matrix to calculate the vicinity directional trees in space division. Finally, in order to maximize the benefits of the database space associated with the model simplification and visualization dependent on the hearing presented deals with the design and analysis of different schemes for transforming textures. As a consequence of the results obtained is proposed for the database space two possible schemes.

Author: CALM PUIG REMEI.
Year: 2004.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: SALA D'ACTES DE LA FME.
Place of preparation: U FACULTAT DE MATEMATIQUES I ESTADISTICA SUD.

Author: AROSTEGUI ORDORICA JOSÉ MANUEL.
Year: 2005.
University: NACIONAL DE EDUCACIÓN A DISTANCIA [www.uned.es].
Place of defense: E.T.S. INGENIEROS INDUSTRIALES.

Summary: This thesis is general approach the development of a two-dimensional model of lignocellulosic biomass pyrolysis. This model includes the whole range of physical and chemical phenomena that are involved during the process (transfer of mass, energy, time and chemical kinetics, etc.) in a porous medium changing, and sensitivity the same in different situations geometrical , heat and humidity. It is proposed to transform the kinetic pyrolysis established based on the fact assumed that the generation of char essentially comes from the decomposition of lignin in place. That transformation introduces a noticeable change in the kinetics putting it into play during pyrolysis process, and makes the model to adapt to different operating conditions. The simulations conducted showed a match between the model results and the different situations pyrolysis registered in industrial practice. Results that have been made on the basis of a criticism of kinetic models in the technical literature. The methodology of the model resolution of the system of equations balance of matter and energy, coupled with a series of tests termogravimétricos carried out, has developed a model consisting of the phenomenon of priólisis and open a number of new forms of treatment in the same terms: By studying the kinetics of the chemical itself through pyrolysis kinetic surfaces. The role of the lignin in the model pyrolysis. The thermogravimetric treatment of moisture in the model.

Author: SUAREZ DE LA TORRE NURIA.
Year: 2006.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: SALA D'ACTES FME.
Place of preparation: U FACULTAT DE MATEMATIQUES I ESTADISTICA SUD.

Summary: In the physically-based fluid animation world, one of the most important challenges is the solution of the Navier-Stokes equations. To solve them in an animation framework, the desired method should be fast, catch all the details of the fluid movement (e.g. splash) and maintain some capability to be manipulated by the animator, making possible the creation of scenes even if they are not physically correct. Often using just one method is not a good solution, mainly because of the complicated behaviour of fluids. In this work we present a coupling method that works trying to profit from the advantages of the two most widely used methods in fluid simulation and to avoid their disadvantages. These methods are: - Smoothed Particles Hydrodynamics (SPH): As a Lagrangian approach, the fluid is supposed to be composed by particles. Each of them has its own material characteristics which are determined by means of kernel functions that define their local influence. This method has very high-level detail but it is really slow and computationally expensive, since the behaviour of every particle depends on its neighbours at every moment. - Marker and Cell (MAC): As Eulerian approach, the fluid values (pressure and velocities) are calculated over a meshed simulation domain. The position of the fluid is determined by marker particles, moved according to the velocity field. Although it needs an iterative system to solve for pressures, it can be considered a fast simulation method, but tends to make features smoother than they are, achieving lower level of detail than SPH. Thus, if we face the problem of simulating a big bulk of fluid where the events needing high-level detail happen near the surface (e.g. a swimming pool or a bath) it seems very appropriated to combine both methods. We have done exactly this: we have studied MAC and SPH in detail and constructed our own algorithms for them; then we have coupled them using MAC for the internal part of the fluid (smoother and more regular) and SPH for the rest. This way, we improve the simulation speed, keeping the level of detail of SPH where it is necessary and its flexibility (it is possible to force the desired movement by means of forces acting over individual particles). These characteristics turn this hybrid method into an interesting tool for fluid animation. On the other hand, as natural complement of the central subject of this work, we have also constructed a basic visualization tool applying an existing Moving Least Squares technique for the visualization of point based geometries to the visualization of our fluid, making the necessary adaptations.

Author: SORIANO ASENSI ANTONIO.
Year: 2006.
University: VALENCIA [www.uv.es].
Place of defense: ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA.
Place of preparation: ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA.

Summary: This thesis describes l proceeding to develop a numerical model based on the technique of in the Finite Difference Time Domain (FDTD1) for the study of electromagnetic propagation in planetary atmospheres. The purpose of the model is presented linking profile electrical conductivity of an atmosphere with electromagnetic spectrum of ELF2 measured. This is intended to facilitate the interpretation of results obtained with the sensor PWA3 embarked on the Huygens probe, which arrived at Titan, one of Saturn's moons, in January 2005. Due to the particular geometry of the system analyzed, it was essential to develop a model FDTD in spherical coordinates. As a result of strong spectral overlap between adjacent frequency resonance it has had to rely on spectral estimation techniques to evaluate the resonance frequencies. Since a three-dimensional model of the Earth's atmosphere is too big and slow, has previously developed a model two-dimensional aprovehando symmetry axis of rotation around the North / South. The two-dimensional model is much faster, which facilitated and expedited the process of purification of the model. Dispose of the two models, a two-dimensional and three-dimensional another, it makes a great versatility. Dependiendo de la situación que se pretenda analizar es posible escoger entre la inmediatez en la obtención de resultados del modelo bidimensional, o en la mayor precisión del modelo tridimensional. Once developed the model has shown the validity of the results returned by studying the Earth's atmosphere. For what has been reproduced with the technique FDTD different numerical experiments performed with other techniques. Subsequently, the results obtained are compared with resutlados pilot numerically frequencies Schumann obtained from the literature. The excellent resutlados obtained with the model Earth justify the appropriateness of the technique used for analysis of other planetary atmospheres. In this regard has been used the same technique to develop models of Titan and Mars.

Author: BONET JORDI.
Year: 2006.
University: BARCELONA [www.ub.es].
Place of defense: FACULTAD DE QUÍMICA.
Place of preparation: FACULTAD DE QUÍMICA (UNIVERSIDAD DE BARCELONA).

Summary: Through a systematic study of alternatives available to propose a novel alternative gets an original alternative to the current residual solvent produced by the industry polivinil alcohol. The polivinil alcohol is a polymer interesting properties: non-toxic, biodegradable and non-flammable, but for every ton of polymer obtained the main reaction produces a ton and a half of waste. The thesis is the result of the question of a Spanish industry to revalue such waste to a marketable product, while recovering raw material for the synthesis of more polymer. The alternative most economically advantageous the transesterificación ethanol is not used industrially because of the high amount of azeotropos that inhibit the separation of products reacicón. In the thesis evaluates alternatives and proposes and discusses an original distillation system by changing pressure reagent that gets only two columns carry out the reaction and retrieve products separated with a total conversion of the reactants. To achieve its goal are determined vapor liquid equilibria for mixtures of compounds, methanol, ethanol, methyl acetate and ethyl acetate. In the calculation procedure is used analysis infinity / infinity for the study of currents between units, an original calculation plate to plate center columno until distillate and walk to link the freflujo with the number of plastos and optimization the variables studied by proporcionalidades. The results were verified through rigorous simulation with PromSim Plus.

Author: SALLERAS FREIXES MARC.
Year: 2006.
University: BARCELONA [www.ub.es].
Place of defense: FACULTAD DE FÍSICA.
Place of preparation: UNIVERSIDAD DE BARCELONA.

Summary: The thermal analysis in microelectronics is one of the most important bottlenecks in current devices. The current densities of integration have made it necessary to include the thermal design from an early stage of product design, whereas in the past it was the last step in the process of developing a product. The highest speed processing in microprocessors and miniaturization of sensors and actuators based thermal also have increased heat flux densities in the devices. All these trends imply reliability problems due to the high temperatures achieved. The need to control these flows of heat in microelectronics is reflected in the growing importance of the tools of CAD (computer aided design, computer-aided design) in the thermal design. The main problem of these tools in its long learning curve and the large amount of computational time and resources to solve complex systems. Therefore, we need models of a reduced ability to respond to new technologies capping where several chips are stacked upon one another in one chip. The current standards in compact models are not applicable to distributions of temperature in a complex where there are these "hot spots" (hot spots) and multiple heat sources. In addition, most of these standards are designed to microelectronic devices and not for microsystems (MEMS), where heat fluxes can be much more complex, and so far, they are only able to predict distributions of temperature stationary. This paper aims to develop a general methodology for obtaining reduced order models that can simulate the heat flux and temperature distribution dináma of any device or Micro. These reduced order models are obtained by identification techniques of systems using transistors temperature simulation of the device or means tested. The proposed methodology is able to work with more complex structures with a heat source and obtain independent models of the boundary conditions. Even can be applied boundary conditions time-varying these reduced order models with satisfactory results. Other methodologies with the same motivation are explored and bought with the proposal of this work. Some of these methodologies are related to the identification systems, but others are based on systems theory linearles or genetic algorithms. This thesis is structured into five chapters. The first chapter is an introduction to design thermal microelectronics and MEMS, which are reviewed current trends in microprocessors, power and MEMS devices, and describes devices and thermal problems in the CAD tools for simularlos. Then in the second chapter represent some examples of systems that have problems opened in thermal modeling. The third chapter is a summary of theory where the main concepts are defined and discussed in depth the thermal systems. It also describes the different methodologies for obtaining reduced order models. Finally, the fourth chapter, the methodologies proposed in the previous chapters are developed for practical cases where required to obtain models reduced. The main findings of this study appear in the final chapter.

Author: VIDAL JIMENEZ BARBARA.
Year: 2006.
University: POLITÉCNICA DE CATALUÑA [www.upc.edu].
Place of defense: SALA DE CONFERENCIES DE ETSEIAT.
Place of preparation: ETSEIB, Edifici H LAB. TERMODINAMICA, PLANTA 8 SD.

Summary: In this thesis a detailed numerical simulation of liquid-solid phase change phenomena has been made, because this phenomenology is of great interest in different industrial areas. The simulation done implies problems of nonlinearity, strong couplings and movable interphase. Like a result, only for the simplest configurations analytical tools can be used, whereas to solve the most of interest problems numerical methods are needed. These methods consist in the discretisation of the equations that define the phenomenology in small cells or control volumes. In this study the Finite Volume Method (FVM) has been used for the governing equations discretisation using Cartesian meshes. A displaced mesh is used; this means that the components of the speed vector are calculated in the faces of the control volumes, which allows a correct coupling between the continuity and momentum equations. So the code as the numerical solutions have been properly verified. The code verification consist of verifying that this is free of programming errors and that the behaviour of the numerical schemes is agreed with the theoretical one. For the verification of the numerical solution the Richardson Extrapolation Method or a mesh refinement study have been used. Once the code and the numerical solutions have been properly verified, the final validation of the process simulation is obtained comparing the numerical results with experimental ones. The Gallium melting problem in a square cavity heated by a side has been widely used by investigators with the objective of evaluating the numerical methods used for solving the phase change phenomena. Although this material have the advantages of its thermophysical properties are well established, the phase change temperature is near the room temperature and is a material with a great industrial interest, also presents some disadvantages like having an anisotropic thermal conductivity of the solid phase. However, the great number of experimental works that can been found in the literature, lead us to choose this material for doing a detailed study of the solid-liquid phase change in this work. Although in the literature this problem appears with different configurations, in this thesis we have centered in the study of the case with an aspect ratio (height/width) of 0.5. Different numerical methods exist to solve solid-liquid phase change problems: methods that follow the moving interphase, methods that fix the moving interphase, etc. In this work we have used the Enthalpy Method because it allows us to use a fixed mesh in all the domain, the Stefan condition is imposed implicitly, it allows the coexistence of more than one front of phase change and allows the interphase has a thickness. This case has the singularity of being in a low Prandtl number range. This causes that a turbulent state has been reached for relatively low Rayleigh numbers. This has made think us on the convenience of making a detailed study to determine the transition of permanent regime to oscillating flow and from oscillating flow to chaotic regime. The problems than have been found in the accomplishment of this work have lead us to the use of multiblock method. This method is used considering incompressible flows and moved meshes. We will explain the modifications that have been necessary for using this method in the phase change phenomena. Two approaches have been used: a conservative one and a pressure based one. For obtaining the reference solution has been use the pressure based method, because the conservative method presents discrepancies between the obtained solution with an only subdomain and the obtained with several subdomains. Finally, a parametric study has been done, considering different aspect ratios, boundary conditions and variations 8 of ±10% 329 in the thermophysical properties with respect to the reference case, with the objective to see as these changes affect on the phenomenology that occupies to us.

Author: Narciso Farias Mercedes Elizabeth.
Year: 2006.
University: AUTÓNOMA DE BARCELONA [www.uab.es].
Place of defense: Escuela Técnica Superior de Ingeniería (ETSE).
Place of preparation: Escuela Técnica Superior de Ingeniería (ETSE).

Summary: To address the complexities introduced optimizing logistical problems, in this dissertation is the proposed methodology, based on the formalism of Colored Petri Network, which allows integrate Research Operations, Artificial Intelligence and Simulation, and the same time combines modeling Systems Events Discretos with the simulation, analysis and optimization systems, allowing transform a conceptual model in a simulation model, and a problem of decision in a matter of searching through the exploration of the state space of the problem. To validate the utility of the proposed methodology has been developed two new methods for exploration and reducing the state space (Method Space States and Method of nodes Repetidos), and two methods for reducing the computation time in the exploration state space (Method of Forbidden Roads and Evaluation of Hierarchical Guards in the PRC). All methods developed have been implemented as part of an Automated System Support Decision Making which has been developed to support the proposed methodology, which can be used both to obtain workable solutions to obtain optimal solutions to optimization problems, as long as the requirements of computational time and memory required to produce such solutions do not exceed available resources. It proposes new lines of research that could help improve computational efficiency considerably to obtain optimal or near-optimal solutions to complex logistical problems of the real world through the methodology proposed in this thesis.

Author: ARETXANDIETA BENGOA XABIER.
Year: 2006.
University: MONDRAGÓN UNIBERTSITATEA [www.mondragon.edu].
Place of defense: MONDRAGON GOI ESKOLA POLITEKNIKOA.
Place of preparation: MONDRAGON GOI ESKOLA POLITEKNIKOA MONDRAGON UNIBERTSITATEA.

Summary: Systems based on components are built today of how the system is verified and validated when the system is fully deployed. This thesis presents a methodology for developing these systems where they anticipate the adequacy of the system to build up a design phase. This suggests behavioral models of components based on state machines that esamblan with a special connector called Composition Management Interface, which allows the simulation of the system construction. All of this is put into ejecucción and analyzes the global state of the system to verify and validate the behavior of a system based on components.