Caracterización de sistemas vibroacústicos mediante análisis estadístico de energía

Hoy día nadie discute la importancia de predecir el comportamiento vibroacústico de estructuras (edificios, vehículos aeronaves, satélites). También se ha hecho patente, con el tiempo, que el rango espectral en el que la respuesta es importante se ha desplazado hacia alta frecuencia en prácticamente todos los campos. Esto ha hecho que los métodos de análisis en este rango alto de frecuencias cobren importancia y actualidad. Uno de los métodos más extendidos para este fin es el basado en el Análisis Estadístico de la Energía, SEA. Es un método que ha mostrado proporcionar un buen equilibrio entre potencia de calculo, precisión y fiabilidad. En un SEA el sistema (estructura, cavidades o aire circundante) se modela mediante una matriz de coeficientes que dependen directamente de los factores de pérdidas de las distintas partes del sistema. Formalmente es un método de análisis muy cómodo e intuitivo de manejar cuya mayor dificultad es precisamente la determinación de esos factores de pérdidas. El catálogo de expresiones analíticas o numéricas para su determinación no es suficientemente amplio por lo que normalmente siempre se suele acabar necesitando hacer uso de herramientas experimentales, ya sea para su obtención o la comprobación de los valores utilizados. La determinación experimental tampoco está exenta de problemas, su obtención necesita de configuraciones experimentales grandes y complejas con requisitos que pueden llegar a ser muy exigentes y en las que además, se ven involucrados problemas numéricos relacionados con los valores de los propios factores de pérdidas, el valor relativo entre ellos y las características de las matrices que conforman. Este trabajo estudia la caracterización de sistemas vibroacústicos mediante el análisis estadístico de energía. Se centra en la determinación precisa de los valores de los factores de pérdidas. Dados los problemas que puede presentar un sistema experimental de estas características, en una primera parte se estudia la influencia de todas las magnitudes que intervienen en la determinación de los factores de pérdidas mediante un estudio de incertidumbres relativas, que, por medio de los coeficientes de sensibilidad normalizados, indicará la importancia de cada una de las magnitudes de entrada (esencialmente energías y potencias) en los resultados. De esta parte se obtiene una visión general sobre a qué mensurados se debe prestar más atención, y de qué problemas pueden ser los que más influyan en la falta de estabilidad (o incoherencia) de los resultados. Además, proporciona un modelo de incertidumbres válido para los casos estudiados y ha permitido evaluar el error cometido por algún método utilizado habitualmente para la caracterización de factores de pérdidas como la aproximación a 2 subsistemas En una segunda parte se hace uso de las conclusiones obtenidas en la primera, de forma que el trabajo se orienta en dos direcciones. Una dirigida a la determi nación suficientemente fiel de la potencia de entrada que permita simplificar en lo posible la configuración experimental. Otra basada en un análisis detallado de las propiedades de la matriz que caracteriza un SEA y que conduce a la propuesta de un método para su determinación robusta, basada en un filtrado de Montecarlo y que, además, muestra cómo los problemas numéricos de la matriz SEA pueden no ser tan insalvables como se apunta en la literatura. Por último, además, se plantea una solución al caso en el que no todos los subsistemas en los que se divide el sistema puedan ser excitados. El método propuesto aquí no permite obtener el conjunto completo de coeficientes necesarios para definir un sistema, pero el solo hecho de poder obtener conjuntos parciales ya es un avance importante y, sobre todo, abre la puerta al desarrollo de métodos que permitan relajar de forma importante las exigencias que la determinación experimental de matrices SEA tiene. ABSTRACT At present there is an agreement about the importance to predict the vibroacoustic response of structures (buildings, vehicles, aircrafts, satellites, etc.). In addition, there has become clear over the time that the frequency range over which the response is important has been shift to higher frequencies in almost all the engineering fields. As a consequence, the numerical methods for high frequency analysis have increase in importance. One the most widespread methods for this type of analysis is the one based on the Statistical Energy Analysis, SEA. This method has shown to provide a good balance among calculation power, accuracy and liability. Within a SEA, a system (structure, cavities, surrounding air) is modeled by a coefficients matrix that depends directly on the loss factors of the different parts of the system. Formally, SEA is a very handy and intuitive analysis method whose greatest handicap is precisely the determination of the loss factors. The existing set of analytical or numerical equations to obtain the loss factor values is not enough, so that usually it is necessary to use experimental techniques whether it is to its determination to to check the estimated values by another mean. The experimental determination presents drawbacks, as well. To obtain them great and complex experimental setups are needed including requirements that can be very demanding including numerical problems related to the values of the loss factors themselves, their relative value and the characteristics of the matrices they define. The present work studies the characterization of vibroacousti systems by this SEA method. It focuses on the accurate determination of the loss factors values. Given all the problems an experimental setup of these characteristics can show, the work is divided in two parts. In the first part, the influence of all the quantities involved on the determination of the loss factors is studied by a relative uncertainty estimation, which, by means of the normalised sensitivity coefficients, will provide an insight about the importance of every input quantities (energies and input powers, mainly) on the final result. Besides, this part, gives an uncertainty model that has allowed assessing the error of one of the methods more widely used to characterize the loss factors: the 2-subsystem approach. In the second part, use of the former conclusions is used. An equation for the input power into the subsystems is proposed. This equation allows simplifying the experimental setup without changing the liability of the test. A detailed study of the SEA matrix properties leads to propose a robust determination method of this SEA matrix by a Monte Carlo filtering. In turn, this new method shows how the numerical problems of the SEA matrix can be overcome Finally, a solution is proposed for the case where not all the subsystems are excited. The method proposed do not allows obtaining the whole set of coefficients of the SEA matrix, but the simple fact of getting partial sets of loss factors is a significant advance and, over all, it opens the door to the development of new methods that loosen the requirements that an experimental determination of a SEA matrix have.

Contribución al análisis cinemático y dinámico de manipuladores paralelos

El documento presentado contiene una aproximación a algunos de los diversos problemas actuales existentes en el campo de la robótica paralela. Primeramente se hace una propuesta para el cálculo de los parámetros estructurales de los robots paralelos, mediante el desarrollo de una metodología que combina las herramientas del estudio de mecanismos con el álgebra lineal; en una segunda sección se propone la solución del problema geométrico directo a partir de la definición de ecuaciones de restricción y su respectiva solución usando métodos numéricos, así como la solución para el problema geométrico inverso; en la tercera parte se aborda el problema dinámico tanto directo como inverso y su solución a partir de una metodología basada en el método de Kane o de trabajos virtuales. Para las propuestas metodológicas expuestas se han desarrollado ejemplos de aplicación tanto teóricos como prácticos (simulaciones y pruebas físicas), donde se demuestra su alcance y desempeño, mediante su utilización en múltiples configuraciones para manipuladores paralelos, entre los que se destacan la plataforma Stewart Gough, y el 3-RRR. Todo con el objetivo de extender su aplicación en futuros trabajos de investigación en el área. ABSTRACT The document presented below provides an approach to some of the many current problems existing in the field of parallel robotics. First is maked a proposal for calculating the structural parameters of the parallel robots, through developing a methodology that combines tools to study mechanisms with the linear algebra; a second section contains a direct geometrical problem solution from the definition of constraint equations and their respective solution using numerical methods, as well as the solution to the inverse geometric problem; in the third part, both, direct and inverse dynamic problem and its solution based on methodology Kane or the method of virtual work are propossed. For each of the exposed methodological proposals they were developed examples of both theoretical and practical application (simulations and physical tests), where its scope and performance is demonstrated by its use in multiple configurations for parallel manipulators, among which stand out the platform Stewart Gough, and 3-RRR. All with the goal of extending its application in future research in the area.

The segmental approximation in multijunction solar cells

The segmental approach has been considered to analyze dark and light I-V curves. The photovoltaic (PV) dependence of the open-circuit voltage (Voc), the maximum power point voltage (Vm), the efficiency (?) on the photogenerated current (Jg), or on the sunlight concentration ratio (X), are analyzed, as well as other photovoltaic characteristics of multijunction solar cells. The characteristics being analyzed are split into monoexponential (linear in the semilogarithmic scale) portions, each of which is characterized by a definite value of the ideality factor A and preexponential current J0. The monoexponentiality ensures advantages, since at many steps of the analysis, one can use the analytical dependences instead of numerical methods. In this work, an experimental procedure for obtaining the necessary parameters has been proposed, and an analysis of GaInP/GaInAs/Ge triple-junction solar cell characteristics has been carried out. It has been shown that up to the sunlight concentration ratios, at which the efficiency maximum is achieved, the results of calculation of dark and light I-V curves by the segmental method fit well with the experimental data. An important consequence of this work is the feasibility of acquiring the resistanceless dark and light I-V curves, which can be used for obtaining the I-V curves characterizing the losses in the transport part of a solar cell.

Comparison of experimental and numerical sloshing loads in partially filled tanks

Sloshing describes the movement of liquids inside partially filled tanks, generating dynamic loads on the tank structure. The resulting impact pressures are of great importance in assessing structural strength, and their correct evaluation still represents a challenge for the designer due to the high level of nonlinearities involved, with complex free surface deformations, violent impact phenomena and influence of air trapping. In the present paper, a set of two-dimensional cases, for which experimental results are available, is considered to assess the merits and shortcomings of different numerical methods for sloshing evaluation, namely two commercial RANS solvers (FLOW-3D and LS-DYNA), and two academic software (Smoothed Particle Hydrodynamics and RANS). Impact pressures at various critical locations and global moment induced by water motion in a partially filled rectangular tank, subject to a simple harmonic rolling motion, are evaluated and predictions are compared with experimental measurements. 2012 Copyright Taylor and Francis Group, LLC.

FE–BE Coupling Methods for Elastoplasticity

Non-linear physical systems of infinite extent are conveniently modelled using FE–BE coupling methods. By the combination of both methods, suitable use of the advantages of each one may be obtained. Several possibilities of FEM–BEM coupling and their performance in some practical cases are discussed in this paper. Parallelizable coupling algorithms based on domain decomposition are developed and compared with the most traditional coupling methods.

Numerical analysis of shell and spatial structures

Since the advent of the computer into the engineering field, the application of the numerical methods to the solution of engineering problems has grown very rapidly. Among the different computer methods of structural analysis the Finite Element (FEM) has been predominantly used. Shells and space structures are very attractive and have been constructed to solve a large variety of functional problems (roofs, industrial building, aqueducts, reservoirs, footings etc). In this type of structures aesthetics, structural efficiency and concept play a very important role. This class of structures can be divided into three main groups, namely continuous (concrete) shells, space frames and tension (fabric, pneumatic, cable etc )structures. In the following only the current applications of the FEM to the analysis of continuous shell structures will be discussed. However, some of the comments on this class of shells can be also applied to some extend to the others, but obviously specific computational problems will be restricted to the continuous shells. Different aspects, such as, the type of elements,input-output computational techniques etc, of the analysis of shells by the FEM will be described below. Clearly, the improvements and developments occurring in general for the FEM since its first appearance in the fifties have had a significative impact on the particular class of structures under discussion.

Análisis teórico de la variación aleatoria de los esfuerzos de montaje de los cables en los puentes atirantados

Los puentes atirantados son una de las tipologías estructurales con mas desarrollo en los últimos años ya que cuentan con un ámbito de aplicación que se ha extendido en gran medida, llegando hasta el rango de luces de más de 1000 m. Por otra parte, para el caso de luces medias, este tipo de puentes aporta unas características resistentes y formales que los hacen muy adecuados en gran número de situaciones. Simultaneamente al importante número de realizaciones llevadas a cabo con esta tipología en los últimos años, se ha producido un gran desarrollo del conocimiento teórico de diferentes aspectos técnicos específicos de estos puentes, tanto a nivel de su comportamiento estructural como de la simulación y mejora de sus procesos constructivos. Estos desarrollos se han producido en gran parte gracias al avance en las capacidad de computación disponible hoy en día con los numerosos y cada vez más sofisticados programas comerciales de cálculo estructural, los cuales permiten la realización de análisis que hasta hace poco tiempo eran muy complicados de desarrollar. Una de las principales características de este tipo de estructuras, además de su elevado hiperestatismo, es la importancia del proceso constructivo, y en concreto del proceso de tesado de los cables, a la hora de conseguir alcanzar una situación final con la estructura en servicio que cumpla los requisitos establecidos a priori. Por este motivo se han llevado a cabo bastantes investigaciones orientadas a la optimización de los axiles a aplicar a los cables en el momento de su colocación y en su situación final. Parte de estos análisis se han centrado en particular en la posible influencia de ciertos parámetros geométricos y mecánicos aleatorios sobre los esfuerzos provocados sobre la estructura. Por otra parte, el propio comportamiento resistente de un tirante introduce ciertas incertidumbres en los formatos de seguridad asociados a los esfuerzos generados sobre ellos, en concreto en relación a la dependencia de estas fuerzas respecto de las cargas permanentes y a su consideración a nivel de coeficientes de seguridad asociados a dichas acciones. En vista de estos aspectos particulares se ha procedido a desarrollar la investigación que se expone a continuación, en la que se ha pretendido obtener un conocimiento adicional de algunos de estos aspectos para así poder contrastar la validez de las hipótesis que hoy en día son establecidas por la normativa vigente. De esta forma el planteamiento adoptado en esta tesis ha sido en primer lugar proponer un método novedoso de introducción de errores aleatorios sobre los esfuerzos o deformaciones de tesado de los cables, de tal manera que se pueda aplicar dicha técnica a distintas tipologías de puentes atirantados. Se pretende con ello conocer la influencia de dichos errores sobre el comportamiento de estas estructuras y valorar la magnitud de las desviaciones finalmente creadas respecto de la situación teórica. Una vez conocida la implicación que supone la aleatoriedad aplicada en la colocación de los tirantes, tanto cualitativa como cuantitativamente, se ha procedido a realizar un análisis de las consecuencias respecto de los formatos de seguridad que de estos datos se deducen. Señalar que esta tesis se ha centrado en el caso concreto de puentes atirantados construidos por voladizos sucesivos y con tableros de hormigón, dado el gran numero de puentes ejecutados con estas características, y con el objetivo de concentrar el estudio realizado en una tipología particular, pero de gran aplicación. Cable-stayed bridges represent one of the most developed structural typologies being used recently as it has a span range that can extend beyond 1000m. Furthermore, when it comes to bridges that span to a medium range we could say their shape and resistance are features that make this kind of bridge suitable for many different applications. Moreover, due to this type of bridge being used newly more extensively it has allowed for a better understanding of the theoretical knowledge of the various technical aspects both at a structural and simulation level in order to improve the constructions process of this type of bridge. Knowledge and development of these bridges can be attributed to the increasing apprehension of I.T. skills and the development of more advanced computer software with regards to structural calculations. The latter can aide more demanding analysis that was previously difficult to ascertain Defining features of this type of structure, besides its hyper static attributes, is the importance of the construction process, namely the cable tensioning process, which determines whether the requirements set out in the construction process will be met once the bridge is finished and traffic loads are applied. This is why much research has been conducted into the optimization of the axial forces to be applied to the cables. Focus, of part of this research, goes into the possible influences of random geometrical and mechanical parametres on the forces applied to the structure. We have therefore proceeded to develop research in which we have tried to gain more in depth knowledge which considers these aspects so that we can validate the hypothesis which are currently established in the regulations Firstly, the purpose of this thesis is to provide a new method to introduce random errors on the forces or deformations of cable tensioning so that this technique can be extended to various models of cable-staying bridges. It is important to highlight that this thesis has focused on cable-stayed bridges built through the balanced cantilever method and with concrete girders and considering there are many actual bridges with these characteristics the research has focused on a specific yet extensively applied method.

WCSPH viscosity diffusion processes in vortex flows

The aim of this paper is to clarify the role played by the most commonly used viscous terms in simulating viscous laminar flows using the weakly compressible approach in the context of smooth particle hydrodynamics (WCSPH). To achieve this, Takeda et al. (Prog. Theor. Phys. 1994; 92(5):939–960), Morris et al. (J. Comput. Phys. 1997; 136:214–226) and Monaghan–Cleary–Gingold's (Appl. Math. Model. 1998; 22(12):981–993; Monthly Notices of the Royal Astronomical Society 2005; 365:199–213) viscous terms will be analysed, discussing their origins, structures and conservation properties. Their performance will be monitored with canonical flows of which related viscosity phenomena are well understood, and in which boundary effects are not relevant. Following the validation process of three previously published examples, two vortex flows of engineering importance have been studied. First, an isolated Lamb–Oseen vortex evolution where viscous effects are dominant and second, a pair of co-rotating vortices in which viscous effects are combined with transport phenomena. The corresponding SPH solutions have been compared to finite-element numerical solutions. The SPH viscosity model's behaviour in modelling the viscosity related effects for these canonical flows is adequate

Cargas explosivas sobre pilares y forjados modelados con elementos tipo viga y lámina

Los fenómenos de impacto y explosión sobre estructuras de hormigón tienen efectos en muchos casos catastróficos a pesar de su reducida probabilidad. Las estructuras de hormigón no suelen estar diseñadas para resistir este tipo de solicitaciones dinámicas. El análisis numérico mediante elementos finitos con integración explícita permite una aproximación suficiente a los efectos de la onda explosiva sobre pilares y forjados de estructuras reticuladas de hormigón. Los materiales recientemente implementados en LS-Dyna para hormigón como el CSCM [1], para elementos de continuo 3D, y la formulación que proporciona la debida compatibilidad con los elementos viga de acero dispuestos de forma segregada, permite estudiar de forma realista modelos detallados de pilares y forjados. Pero las limitaciones computacionales hacen inviable emplear estos métodos en estructuras completas. Como alternativa es posible usar modelos de elementos estructurales de vigas y láminas para el análisis de estas estructuras. Sin embargo es necesario un adecuado ajuste de parámetros y propiedades en estos modelos. Este trabajo muestra un método con en el que obtener modelos de elementos estructurales, elementos viga y lámina, usando modelos de material [2] adecuados para ellos, junto a un procedimiento para incluir la armadura de forma adecuada. Utilizando este método es posible representar con suficiente aproximación el comportamiento de modelos detallados realistas de forjados y pilares de estructuras reticuladas de hormigón frente a acciones explosivas, posibilitando el análisis de una estructura completa frente a explosión.

Análisis de los parámetros que definen un modelo de daño aplicado al deterioro de la capacidad resistente de paredes arteriales

El comportamiento mecánico de las paredes arteriales constituye un ingrediente fundamental para entender enfermedades cardiovasculares como las disecciones aórticas o la fisuración de placas de ateroma. El fenómeno mecánico que subyace en estas enfermedades es el fallo constitutivo del material, que puede deberse a la degradación de las propiedades mecánicas del tejido, a la aplicación de una carga excesiva o la concentración de tensiones que tiene lugar en traumatizaciones de la íntima. Un modelo que busque reproducir este comportamiento biomecánico debe incorporar además la información histológica de las paredes arteriales, formadas en una primera aproximación por una matriz acuosa incompresible y dos familias de fibras dispuestas helicoidalmente. De entre los modelos mecánicos existentes cabe destacar los modelos de daño, que con un conjunto pequeño de parámetros son capaces de reproducir la degradación de la capacidad resistente tanto en la matriz como en las fibras. Esta simplicidad convierte a esta clase de modelos en una herramienta muy útil para estudiar este fenómeno. En este trabajo se presenta un estudio para un modelo de daño de los parámetros que lo definen, asociados a la imposición de la incompresibilidad de la matriz acuosa y a las evoluciones del daño en dicha matriz y en las dos familias de fibras. Se incluye una revisión de las técnicas existentes para imponer la incompresibilidad, un resumen bibliográfico de la caracterización experimental de los parámetros materiales y un análisis de su influencia en el comportamiento estructural de la aorta, analizada mediante ensayos de tracción

Dynamic Effects On Culverts for High Speed Trains

Culverts are very common in recent railway lines. Wild life corridors and drainage conducts often fall in this category of partially buried structures. Their dynamic behavior has received far less attention than other structures such as bridges but its large number makes that study an interesting challenge from the point of view of safety and savings. In this paper a complete study of a culvert, including on-site measurements as well as numerical modelling, will be presented. The structure belongs to the high speed railway line linking Segovia and Valladolid, in Spain. The line was opened to traffic in 2004. Its dimensions (3x3m) are the most frequent along the line. Other factors such as reduced overburden (0.6m) and an almost right angle with the track axis make it an interesting example to extract generalized conclusions. On site measurements have been performed in the structure recording the dynamic response at selected points of the structure during the passage of high speed trains at speeds ranging between 200 and 300km/h. The measurements by themselves provide a good insight into the main features of the dynamic behaviour of the structure. A 3D finite element model of the structure, representing its key features was also studied as it allows further understanding of the dynamic response to the train loads . In the paper the discrepancies between predicted and measured vibration levels will be analyzed and some advices on numerical modelling will be proposed

Towards the foundation of a global modes concept

A contribution is presented, intended to provide theoretical foundations for the ongoing efforts to employ global instability theory for the analysis of the classic boundary-layer flow, and address the associated issue of appropriate inflow/outflow boundary conditions to close the PDE-based global eigenvalue problem in open flows. Starting from a theoretically clean and numerically simple application, in which results are also known analytically and thus serve as a guidance for the assessment of the performance of the numerical methods employed herein, a sequence of issues is systematically built into the target application, until we arrive at one representative of open systems whose instability is presently addressed by global linear theory applied to open flows, the latter application being neither tractable theoretically nor straightforward to solve by numerical means. Experience gained along the way is documented. It regards quantification of the depar- ture of the numerical solution from the analytical one in the simple problem, the generation of numerical boundary layers at artificially truncated boundaries, no matter how far the latter are placed from the region of highest flow gradients and, ultimately the impracti- cally large number of (direct and adjoint) modes necessary to project an arbitrary initial perturbation and follow its temporal evolution by a global analysis approach, a finding which may question the purported robustness reported in the literature of the recovery of optimal perturbations as part of global analyses yielding under-resolved eigenspectra.

Modelling Crowd Load for Floor Vibration Analysis

The dynamic floor loads induced by crowds in gymnasium or stadium structures are commonly modelled by superposition of the individual contributions using reduction factors for the different Fourier coefficients. These Fourier coefficients and the reduction factors are calculated using full scale measurements. Generally the testing is performed on platforms or structures that can be considered rigid, such that the natural frequencies are higher than the frequencies of the spectator movement. In this paper we shall present the testing done on a structure that used to be a gymnasium as well as the procedure used to identify its dynamic properties and a first evaluation of the socalled “group effect”.

Statistical mechanics approaches to granular media: between micromechanics and macromechanics

The mechanical behavior of granular materials has been traditionally approached through two theoretical and computational frameworks: macromechanics and micromechanics. Macromechanics focuses on continuum based models. In consequence it is assumed that the matter in the granular material is homogeneous and continuously distributed over its volume so that the smallest element cut from the body possesses the same physical properties as the body. In particular, it has some equivalent mechanical properties, represented by complex and non-linear constitutive relationships. Engineering problems are usually solved using computational methods such as FEM or FDM. On the other hand, micromechanics is the analysis of heterogeneous materials on the level of their individual constituents. In granular materials, if the properties of particles are known, a micromechanical approach can lead to a predictive response of the whole heterogeneous material. Two classes of numerical techniques can be differentiated: computational micromechanics, which consists on applying continuum mechanics on each of the phases of a representative volume element and then solving numerically the equations, and atomistic methods (DEM), which consist on applying rigid body dynamics together with interaction potentials to the particles. Statistical mechanics approaches arise between micro and macromechanics. It tries to state which the expected macroscopic properties of a granular system are, by starting from a micromechanical analysis of the features of the particles and the interactions. The main objective of this paper is to introduce this approach.

A C1 finite element for Kirchhoff plate bending

After a short introduction the possibilities and limitations of polynomial simple elements with C1 continuity are discussed with reference to plate bending analysis. A family of this kind of elements is presented.. These elements are applied to simple cases in order to assess their computational efficiency. Finally some conclusions are shown, and future research is also proposed.