Search

This paper presents a finite element solver for the simulation of steady non-Newtonian flow problems, using a regularized Bingham model, with adaptive mesh refinement capabilities. The solver is based on a stabilized formulation derived from the variational multiscale framework. This choice allows the introduction of an a posteriori error indicator based on the small scale part of the solution, which is used to drive a mesh refinement procedure based on element subdivision. This approach applied to the solution of a series of benchmark examples, which allow us to validate the formulation and assess its capabilities to model 2D and 3D non-Newtonian flows.

The maturity of numerical techniques both in Computational Structural and Fluid Dynamics, accompanied by the availability of affordable high-performance hardware, allows the use of virtual rapid-prototyping facilities as realistic alternatives to traditional wind tunnel experiments. This work aims to present a virtual wind tunnel tool that combines both “standard” body-fitted approaches and innovative embedded techniques. Two are the distinctive features of the work. Firstly, it allows the simulation of membrane structures subjected to large displacements. This is typically challenging both for embedded solvers, due to the need of modelling a strong discontinuity of the flow field in correspondence of the structural position, and for body-fitted techniques, due to the large displacement regime. Secondly, it allows combining body fitted and embedded structures within the same simulation. All the presented strategies are developed both in Open MP and MPI.

In recent years the technology on embankments dams has developed sensibly due to the advances in soil mechanic, and in all related sciences. Nevertheless their vulnerability to overtopping still remains their weakest point in comparison with concrete structures. The principal aim of this theme is the simulation of the initial stage of failure of the dam when an overtopping or an exceptional flood occurs. A fluid-structure coupled problem has to be considered. The sudden variation of the upstream conditions induces a quick evolution of a seepage line in the downstream shoulder. Non linear Darcy law has to be taken into account. On the other hand the water, emerging from the toe of the dam, induces dragging of particles and possible mass sliding, depending on geometrical and material conditions. UPM and CEDEX have carried out more than 70 experiments during the last two years. They analyzed the influence of a series of parameters on the failure mechanisms. These parameters are, for instance, the dimension of the rocks, the slope of the downstream part of the dam, the type of impermeable element used, and so on. We propose to reproduce numerically three experiments analyzing the evolution of seepage and following initial stage of failure in the case of a homogeneous dam, of a dam with impermeable upstream face and of a dam with internal core. Experimental data, bottom pressure distribution and topographic analysis of the geometry of the dam during the failure can be compared with numerical results.

This work presents a model which allows considering the interaction between rockfill dams and the water. The fluid behaviour is analyzed on an Eulerian fixed mesh. A level set technique is employed to tack the evolution of the free surface. An edge based approach is used to solve efficiently the modified form of the Navier‐Stokes equations. The dam response is evaluated on a Lagrangian moving mesh using PFEM. A visco‐rigid constitutive model is used to describe the structural behaviour and collapse of rockfill under fluid dynamic forces. Mohr‐Coulomb is the failure criteria adopted in the calculation. Finally a nodal projection algorithm allows the transferring of information through non matching meshes.

In recent years the technology on embankments dams has developed sensibly due to the advances in soil mechanic, and in all related sciences. Nevertheless their vulnerability to overtopping is still an open theme of discussion. Such an extreme phenomenon can lead to catastrophic consequences with loss of stability and failure of the entire structure. In spite of that, the consequences of an overtopping are still difficult to analyze and to prevent optimally. The possibility to define a numerical instrument to provide support for analyzing the failure of a dam is an important step ahead for organizing the intervention measures and for optimizing the economic plan. This is the principal reason why the Spanish ministry of Science and Innovation is funding the XPRES project, between the Polytechnic University of Madrid (UPM), the Centre for Hydrographical Studies of CEDEX and the International Centre for Numerical Methods in Engineering (CIMNE). The characterization of the failure process is dealt with by combining new numerical models, developed by CIMNE, with experimental tests on physical models of different sizes and scales carried out by UPM and CEDEX. In the present paper the work carried out by CIMNE is presented. The numerical model is based on combination of Lagrangian Particle Finite Element Method (PFEM) and Eulerian level set techniques in order to optimize the coupling between the dynamic effect of water inside and outside the dam, with the structural deformation and collapse induced on the down stream shoulder. The breaching formation and the consequent beginning of the failure are analyzed with particular interest. The specific features of the adopted method make it appropriate to treat the rockfill material and its large deformations and shape changes. A variable viscosity law and an energetic erosion criterion are used to take into account mass sliding and erosion acting in combined or alternative way.

The present work is concerned with some recent advances in FSI simulations with particular emphasis on the simulation of rockfill dams during extreme phenomena. The final objective is the study of the consequences of an overtopping on the stability of such structure and its possible failure mode. A mixed Lagrangian and Eulerian approach is used. The fluid behavior is described using a modified form of the Navier Stokes equations in order to consider the effect of a variable porosity. A non linear Darcy law is included in the momentum equation. A level set function is chosen to follow the movement of the free surface inside and outside the porous medium. The structure is described using a purely lagrangian PFEM formulation. The specific features of PFEM make it appropriate to treat the rockfill material and its large deformations and shape changes. A projection technique allows to perform the data transfer between the fluid and the structure non matching meshes.

In order to validate the Xfinas code a very comprehensive series of test examples were solved by Prof Ki-Du Kim and his co-workers. A collection of the more representative benchmarks were chosen at CIMNE for testing the good behavior of every element implemented in the software. The aim of the validation work carried out at CIMNE has been to asses the accuracy of the Xfinas program. This was done studying the whole validation process carried out by Prof Ki-Du Kim’s team in detail. For this purpose we have chosen at CIMNE randomly the different benchmarks to be reproduced between those of the validation manual (VM from now on). In every example we checked the agreement of the results with the Xfinas validation data.

An increasing interest is rising on the study of rockfill embankment dams during overspilling. Design criteria of earth dams are being reviewed in many countries to guarantee an increasing safety level in front of an exceptional flooding. The possibility of studying the failure process is currently limited by the lack of a precise knowledge and by the absence of a suitable computational method which can predict accurately the onset and evolution of breaching. The analysis of the possible consequences of an accidental overspill is then impossible or very imprecise and the necessary economical measures for solving the problem are then inefficient. An appropriate computational method will help to reduce the economic impact of the investments in dam safety and in emergency plans for embankment dams. The dam failure is mainly due to the hydrodynamic forces acting on the rockfill material in the downstream face. Two failure mechanisms occur in an alternative or combined mode: dragging of rockfill particles with erosive effects, and loss of stability of a part of the downstream region due to the slippage of a bulk mass of the rockfill material. The loss of a relevant part of the rockfill material in the downstream side of the dam, can lead to mechanical failure of the impervious region, either if this is formed by a concrete (or asphalt) face, or by a clay core. In the latter case, erosion of the cohesive material can also influence the failure process, but the scale time of appearance of this erosion phenomenon is so different from the principal erosion of the rockfill part, that this second one is not taken into account. The objectives of the current work is to develop and validate a new computational method of general applicability that allows treating above problems. The method will combine advanced finite element and particle techniques. Fluid-structure interaction effects and non-linear geometrical and mechanical effects in the dam material will be considered. The specific features of the analysis method proposed make it appropriate to treat the rockfill material adjacent to the downstream surface where a transfer of momentum between the water running over the dam surface and that moving through the interior due to seepage occurs

Several comparisons between experiments and computational models are presented in the following pages. The objective is to verify the ability of Particle Finite Elements Methods (PFEM) [1] [2] to reproduce hydraulic phenomena involving large deformation of the fluid domain [4].

Bulk handling, transport and processing of granular materials and powders are fundamental operations in a wide range of industrial processes and geophysical phenomena. Particulate materials, which can be found in nature, are usually characterized by grain size which can range across several scales: from nanometre to the order of metre. Depending on the volume fraction and shear strain conditions, granular materials can have different behaviours and often can be expressed as a new state of matter with properties of solids, liquids and gases. For the above reasons both the experimental and the numerical analysis of granular media is still a difficult task and the prediction of their dynamic behaviour still represents nowadays an important challenge. The main goal of the current thesis is the development of a numerical strategy with the objective of studying the macroscopic behaviour of dry granular flows in quasi-static and dense flow regime. The problem is defined in a continuum mechanics framework and the balance laws, which govern the behaviour of a solid body, are solved by using a Lagrangian formalism. The Material Point Method (MPM), a particle-based method, is chosen due to its features which make it very suitable for the solution of large deformation problems involving complex history-dependent constitutive laws. An irreducible formulation using a Mohr-Coulomb constitutive law, which takes into account geometric non-linearities, is implemented within the MPM framework. The numerical strategy is verified and validated against several benchmark tests and experimental results, available in the literature. Further, a mixed formulation is implemented for the solution of granular flows that undergo undrained conditions. Finally, the developed MPM strategy is used and tested against the experimental study performed for the characterization of the flowability of several types of sucrose. The capabilities and limitations of this numerical strategy are observed and discussed and the bases for future research are outlined. El manejo, el transporte y el procesamiento de materiales granulares y polvo son operaciones fundamentales en una amplia gama de procesos industriales y de fenómenos geofísicos. Los materiales particulados, que pueden ser encontrados en la naturaleza, generalmente están caracterizados por el tamaño del grano, que puede variar entre varios órdenes de magnitud: desde el nanómetro hasta el orden de los metros. En función de las condiciones de fracción volumétrica y de deformación de cortante, los materiales granulares pueden tener un comportamiento diferente y a menudo pueden expresarse como un nuevo estado de materia con propiedades de sólidos, de líquidos y de gases. A causa de las observaciones antes mencionadas, tanto el análisis experimental como la simulación numérica de medios granulares es aún una tarea compleja y la predicción de su comportamiento dinámico representa aun hoy día un desafío muy importante. El principal objetivo de esta tesis es el desarrollo de una estrategia numérica con la finalidad de estudiar el comportamiento macroscópico de los flujos de medios granulares secos en régimen cuasiestático y en régimen dinámico. El problema está definido en el contexto de la mecánica de medios continuos y las leyes de equilibrio, que gobiernan el comportamiento del cuerpo sólido, y están resueltas mediante un formalismo Lagrangiano. El Metodo de los Puntos Materiales (MPM), método basado en el concepto de discretización del cuerpo sólido en partículas, está elegido por sus características que lo convierten en una técnica apropiada para resolver problemas de grandes deformaciones donde se tienen que utilizar complejas leyes constitutivas. En el marco del MPM está implementada una formulación irreducible que usa una ley constitutiva de Mohr-Coulomb y que tiene en cuenta no-linealidades geométricas. La estrategia numérica está verificada y validada con respecto a tests de referencia y resultados experimentales disponibles en la literatura. Además, se ha implementado una formulación mixta para resolver casos de flujo granular en condiciones no drenadas. Por último, la estrategia MPM desarrollada está utilizada y evaluada con respecto a un estudio experimental realizado para la caracterización de la fluidez de diferentes tipologías de azúcar. También se presentan unas observaciones y discusión sobre las capacidades y las limitaciones de esta herramienta numérica y se describen las bases de una investigación futura.

Heat treatment processing consists in heating a metallic material above austenitization temperature to later quickly cool it down. The accurate simulation of this phenomenon is a current industrial need in order to predict the outcome of their processes and therefore save time and money avoiding extensive prototyping phases. The aim of this master thesis is to integrate a microstructural model inside the commercial code Stampack, such that it can be used in simulation of hot stamping, in which a blank sheet is quenched and shaped at the same time

Nowadays, dam safety is becoming more relevant in our society due to the importance of its functions (power generation, water supply, flood control) and the severity of the consequences in case of a serious breakdown. The "safety of hydraulic infrastructures" is one of the priorities according to the R&D Spanish State Program which is oriented to social challenges. Thanks to recent improvements in the modelling of dams as well as the evolution of computational resources, it is possible to perform more detailed analysis. However, most of commercial software is devoted to more common problems (e.g. buildings, bridges) without considering specific aspects of dams: concrete ageing process, joints during the construction process, contact with the ground or uplift pressure. The development of a specific application for dam engineering, both for construction phase and operating period, is the main objective of this project. The thesis presents the computational tool, which is based on finite element formulations and solves thermomechanical problem using weak coupling, designed for analysing the operating period. The proposed software is validated with a real case: La Baells dam. The computational results have shown excellent agreement with the obtained data during monitoring process.

Implementation of a FSI procedure involving the interaction of shell structures and fluids.

The meshless method Smoothed Particle Hydrodynamics (SPH) is one of the primarily developed numerical methods in the meshless methods family. It has been proven to work well with problems of free surface fluid flows. For the purpose of this thesis, the Smoothed Particle Hydrodynamics method and its current capabilities were studied. Using the obtained knowledge from the literature, a free surface fluid flow application was implemented in the framework of KRATOS multiphysics. Some well known benchmark problems were reproduced using the generated SPH solver. The algorithm's ability to catch the correct free surface evolution and obtain an accurate pressure distribution of free surface problems were analyzed. Furthermore, some programming aspects of coding for meshless methods were discussed from the computational workload point of view. Using the developed application, it was demonstrated that in order to create a generic SPH fluid solver that is able to model any type of fluid, some additional corrections must be applied to the SPH solution. These corrections, although being very convenient to obtain good particle tracking of the fluid motion, affect the pressure distribution of the flow excessively and make the weakly compressible SPH method untrusted for the pressure results of fluid flow problems.