Seminars

IMMC

The institute of Mechanics, Materials and Civil Engineering (iMMC) organizes a series of seminars about several themes :

  • Energy, thermodynamics, and chemical engineering
  • Mechatronics and biomechanics
  • Processing, materials, assembly, and structures
  • Fluid mechanics, environmental engineering, and geomechanics
  • Numerical and computational methods
Here follows the list of seminars.



Seminars to come

Previous seminars

30/08/2017(14:00) [Lieu : BARB93 Auditorium]
Auke Ijspeert(EPFL)
"Neuromechanical models of locomotion: from biology to robotics"
The ability to efficiently move in complex environments is a fundamental property both for animals and for robots, and the problem of locomotion and movement control is an area in which neuroscience and robotics can fruitfully interact. Animal locomotion control is in a large part based on spinal cord circuits that combine reflex loops and central pattern generators (CPGs), i.e. neural networks capable of producing complex rhythmic or discrete patterns while being activated and modulated by relatively simple control signals. These networks are located in the spinal cord for vertebrate animals and interact with the musculoskeletal system to provide "motor primitives" for higher parts of the brain, i.e. building blocks of motor control that can be activated and combined to generate rich movements. In this talk, I will present how we model the spinal cord circuits of lower vertebrates (lamprey and salamander) using systems of coupled oscillators, and how we test these models on board of amphibious robots. The models and robots were instrumental in testing some novel hypotheses concerning the mechanisms of gait transition, sensory feedback integration, and generation of rich motor skills in vertebrate animals. I will also discuss how the models can be extended to control biped locomotion and lower-limb exoskeletons.


16/06/2017(16:15) [Lieu : Stévin - b. (-145)]
Quentin Docquier(iMMC / MEED)
"Optimal Control of the Tilt Control for a Narrow Track Vehicle through MBS Modelling"
Narrow track vehicles, characterized by a large height-to-track ratio, need to be leaned toward the inside of a turn to prevent them from overturning in case of high-speed cornering. Two modes have been proposed for active tilt control. The Direct Tilt Control (DTC) consists in forcing the tilting of the chassis with an actuator in reaction to the pilot steering. Conversely, in Steering Tilt Control (STC), the steering wheel is decoupled from the wheels. An actuator steers the wheels to lean the vehicle in an appropriate angle which consequently induces a desired turn (i.e. a steering wheel orientation given by the pilot). In this work, the relevance of using both control modes in a combined tilt strategy SDTC (Steering and Direct Tilt Control) is studied. A model for a Narrow Track vehicle has been developed based on a multibody formalism (through MBS software Robotran). The model features an extra rotational degree of freedom which allows the chassis to tilt with respect to the lower part of the vehicle. The overall complexity of the vehicle and its non-linearity are taken into account in the derivation of the equations of motion. In addition, an optimization code is implemented with the help of CasADi. With this tool, Optimal Control Problems are solved in order to determine which actuator (steering and/or tilting) should be used to track given trajectories for the vehicle lean angle. It has been showed that a combined strategy (SDTC) does not yield significant improvement in terms of energy consumption compared to the steering tilt control (STC) alone. Similar results are observed while model predictive control is considered with again the SDTC performing the same way as the STC does. However, when constraints are applied to limit the vehicle (e.g. limitation of counter-steering), the SDTC yields to the use of the tilting actuator for the initiation of the vehicle lean. In addition, the tool developed already allows to study Optimal Control law applied to multibody systems.


09/06/2017(1:00 - 1:55) [Lieu : Euler seminar room]
Jérémy Chevalier(iMMC / IMAP)
"Unveiling the heterogeneity controlled mechanical behavior of epoxy resins"
Highly cross-linked epoxies constitute the materials of choice used as matrix in high performance glass or carbon reinforced composites, dictating a significant part of their mechanical behavior. As for thermoplastics, their mechanical response involves a complex intermingling of elasticity, viscoplastic yielding with softening and large strain hardening, creep and non-linear recovery upon unloading. Continuum viscoelastic viscoplastic constitutive models have been developed to capture all these features but at the price of introducing typically more than twenty phenomenological parameters, and often many more. Here a mesoscale model based on the heterogeneous transformation of molecular shear transformation zones (STZ), relying on only five parameters to be identified, is shown to be rich enough to quantitatively reproduce all these effects without relying on any viscoelastic or plastic ad hoc assumptions. The model is identified and validated based on a large set of experimental data obtained on the RTM6 epoxy resin, at different rates, temperature under loading and unloading conditions, involving creep and recovery tests. The key element is the generation of large back stress due to the random thermo-mechanical activation of the STZ which eventually combine into collective topological interactions, allowing, among others, to predict complex forward and backward creep and recovery upon unloading without any additional fitting parameters. This approach offers an alternative to complex phenomenological models, allows the identification of such models avoiding long and costly experimental programs and can be used to address low scale micromechanical response of small composite unit cells.


02/06/2017(13:00 - 13:55) [Lieu : Euler seminar room]
Quentin De Radiguès( iMMC/IMAP)
"Hydrogen as storage for green electricity: 3D electrodes and process intensification. "
Hydrogen is a promising and well-accepted energy vector to store electricity produced by intermittent sources, such as solar panels and wind turbines. In order to be cost effective, the water electrolysis needs to be intensified (higher production in smaller units) and scaled up to the match the power of electricity sources. This presentation will focus on the importance of storage of electricity and process intensification of water electrolysis. In a second part, applications of hydrogen in power to mobility and power to hydrogen will be presented. At last, I will show the interest of 3D electrodes and the pilot plant scale up aimed in the project WallonHY.


02/06/2017(11 AM) [Lieu : Ba91 auditorium]
Basile Audoly(Ecole Polytechnique)
"The non-linear mechanics of slender deformable bodies"
There are many examples of slender (quasi-one dimensional) bodies around us, including computer cables, human hair, sailing and climbing ropes. Being flexible, slender bodies can undergo large rotations. As a result, they often display complex and interesting behaviors, among which are buckling phenomena. Being effectively one-dimensional, slender bodies are governed by equations that are simpler to write down, more likely to have analytical solutions, and easier to solve numerically than the corresponding equations of 3D continuum mechanics: this allows one to address problems that would be much more difficult (or impossible) to tackle using 3D models. In this talk, I will discuss the derivation of 1D models by dimension reduction, and their analytical or numerical solution in the non-linear regime, based on three specific examples: the coiling of thin viscous threads, the buckling of thin rods with incompatible strain, and the phenomenon of necking localization in elastic bars.


19/05/2017(17:00) [Lieu : Stévin - b. (-145)]
Caroline Leroi(iMMC / MEED)
"Conducted disturbances in the frequency range 2-150 kHz"
In the last decades, the power grid has significantly changed. One of the main change is the rapid growth of devices using power electronics. These devices include switches with frequency commutation ranging between 2 kHz and 150 kHz. Besides, the Power Line Carrier (PLC) is also working in this frequency range. The PLC is a means of communication which uses the existing power network. Thus, there is a coexistence of intentional and the so-called “unintentional” emissions in this frequency band. Consequently, electromagnetic compatibility (EMC) has to be guaranteed. The problem is that the standardization which is supposed to regulate the emission of the perturbations and the immunity of sensitive devices is, currently, almost non-existent for these frequencies. However, from the standardization perspective, closing the gap in this frequency range has become a priority. Compatibility levels, emission limits and immunity limits are needed. A correct setting of these levels is thus of major importance. Indeed, an inappropriate choice could create an unnecessary barrier against the introduction of equipments like solar panels, electric cars, etc. Furthermore, it could also affect the expansion of smart grids. The aim of the research is to contribute to a better understanding of the origin, the propagation and the consequences of emissions in the frequency range 2-150 kHz. It is important to gather knowledge in order to set appropriate limits in the standards. In this seminar, the context of this problematic will be further explained. Results concerning disturbances induced by small wind turbines will be presented. And finally, a first step concerning the modelling of LV cables will be introduced.


17/05/2017(13:00 - 14:00) [Lieu : Réaumur - d.382]
Adeline Delvaux(iMMC / IMAP)
"Électrodes nano-structurées en nickel pour la production électrochimique d’hydrogène"
Dû à l’impact négatif des sources d’énergies actuelles sur l’environnement, nous nous devons de les remplacer par des énergies plus vertes. Néanmoins, le caractère intermittent des énergies renouvelables telles que l’énergie solaire ou éolienne reste un problème à résoudre. Dans ce contexte, l’électrolyse de l’eau à partir d’électricité verte permet le stockage d’énergie sous forme d’hydrogène sans émission toxique pour l’homme ou l’environnement. Ce procédé d’électrolyse de l’eau a déjà été largement étudié, mais l’efficacité des cellules électrochimiques doit encore être améliorée afin de les développer à grande échelle. La performance des électrodes étant l’un des facteurs limitant l’efficacité, nous pouvons accroître le transport de masse lors de l’électrolyse en améliorant le détachement des bulles de gaz formées durant le procédé. Cela est possible en modifiant la morphologie de la surface des électrodes. Des électrodes nano-structurées en nickel sont alors utilisées pour évaluer les améliorations de la performance électrochimique résultant de ces modifications de surface.


12/05/2017(1:00 - 1:55 pm) [Lieu : Euler seminar room]
Raphaël Janssens(iMMC / IMAP)
"Slurry photocatalytic membrane reactor technology for removal of pharmaceutical compounds from wastewater: towards cytostatic drug elimination"
The potential of photocatalytic membrane reactors (PMR) to degrade cytostatic drugs is presented in this work as an emerging technology for wastewater treatment. Cytostatic drugs are pharmaceutical compounds (PhCs) commonly used in cancer treatment. Such compounds and their metabolites, as well as their degraded by-products have genotoxic and mutagenic effects. A major challenge of cytostatic removal stands in the fact that most drugs are delivered to ambulant patients leading to diluted concentration in the municipal waste. Therefore safe strategies should be developed in order to collect and degrade the micro-pollutants using appropriate treatment technologies. Degradation of cytostatic compounds can be achieved with different conventional processes such as chemical oxidation, photolysis or photocatalysis but the treatment performances obtained are lower than the ones observed with slurry PMRs. Therefore the reasons why slurry PMRs may be considered as the next generation technology will be discussed in this work together with the limitations related to the mechanical abrasion of polymeric and ceramic membranes, catalyst suspension and interferences with the water matrix. Furthermore key recommendations are presented in order to develop a renewable energy powered water treatment based on long lifetime materials.


05/05/2017(1:00 - 1:55 pm) [Lieu : Euler seminar room]
Audrey Favache(iMMC/IMAP)
"From plate theory to scratch resistance"
The CARUSO project aims to develop performance indexes to asses and improve the scratch resistance of coatings. This seminar will present some results obtained in this framework in the particular case of thin hard films deposited on compliant substrates. Results of the plate theory have been used to derive performance indexes for substrate yielding and film cracking.


26/04/2017(14:00-15:00) [Lieu : UCL - Bât Euler - a.002 ]
Pierre Anquez(Université de Lorraine)
"Geological surface model repair and simplification"
Geological 3D surface models are built to represent the major subsurface structures of a given area. The challenge is that these models do not meet the prerequisites of 3D meshing methods. Repairing and/or simplifying geological surface models is then a way to bridge the gap between surface model conception and volumetric meshing generation. In this seminar, I will present my ongoing research on geological surface model connectivity updating and simplifying, in order to define watertight rock volumes, and make their volumetric meshing possible.


24/04/2017(14:00-15:00) [Lieu : UCL - Bât Euler - a.002 ]
Célestin Marot(MEMA)
"Parallel Radix Sort"
The idea of Radix Sorting is to sort numbers not by comparing them algebraically, but by sorting them digit by digit, from least significant digit to most significant digit. We will discover the advantages and disadvantages of a parallel radix sort compared to the classical quicksort and characterize the situations where standard qsort and std::sort should be dropped.


21/04/2017(16:15) [Lieu : Stévin - b. (-145)]
Olivier Lantsoght(iMMC/MEED)
"Multibody System and granular media "
The coupling between two different domains is the subject of this research. On the first side we have the Multibody System (MBS) dynamics formalism which is now widely used to understood, predict or control the behavior of a large panel of mechanisms (vehicles, tools, robots). On the other side, we have the analysis of contacts dynamics. Contact is everywhere: between a mechanism and the environment (a wall, a granular media, the ground…), the mechanism with itself, inside the articulation of the mechanism… In the current approach we choose to model the contact with the Non-Smooth Contact Dynamics (NSCD) method introduced in the eighties. This approach has the advantage to require few parameters describing the macro-behavior of the contact such as the friction coefficient, while the apparition of jump in the velocity field is the most challenging task in terms of modelling and simulation. It has been shown that a co-simulation between MBS and NSCD is not suitable due to the high stiffness of the contact. In our approach we solve the equations describing the multibody system, the environment (e.g. granular media) and the contact dynamic together, using a strong coupling approach. In order to validate our numerical development, we introduce a benchmark that represent a multibody mechanism interacting with a vibrated granular media, highlighting the impact of phenomena such as the fluidification of the granular media on the dynamics of the whole system. This benchmark has been designed using our computer model and is now being built to validate the numerical method.


21/04/2017(1:00 - 1:55 pm) [Lieu : Euler seminar room]
Guerric Lemoine(iMMC )
"Localized necking in nanocrystalline metallic thin films"
From a reliability viewpoint, ductility in nanocrystalline metallic thin films is a critical issue in many applications such as in micro- or nano-electromechanical systems, in thin membrane technology and in flexible electronics. An enriched Marciniak-Kuczynski type imperfection model is developed. The model is devoted to localized necking in which the length of the neck scales with the thickness of the specimen. Compared to the earlier analysis of Hutchinson and Neale, devoted on the synergy between imperfection, strain hardening and rate sensitivity, the present study aims at accounting also for strain gradient plasticity effects, for grain size dependent strength and rate sensitivity, for elasticity and for possible contributions of creep mechanisms.


31/03/2017(16:15) [Lieu : Stévin - b. (-145)]
Antoine Falisse(KU Leuven)
"Predictive simulations to explore gait control in children with cerebral palsy and provide guidance in the clinical decision making process"
Cerebral palsy (CP) is the most common cause of physical disability among children in Europe. Children with CP suffer from brain lesions resulting in various impairments undermining their gait pattern and eventually their mobility. Currently, the clinical decision for selecting the treatment that best improves the gait pattern mainly rests on the expert intuition and experience of the clinicians. It is a particularly difficult decision to take as the complex interplay between mechanics and motor control makes the prediction of the treatment outcome very hard. The use of musculoskeletal models to predict the effect of treatment options on gait performance is therefore particularly valuable and increasingly popular. However, the accuracy of state-of-the-art model-based predictions is still limited mainly due to the non-specificity of the models, the high computational costs associated with the predictive simulations, and our poor knowledge of gait control strategies. In my PhD project, I aim to address these limitations to improve the predictions. In this seminar, I will first present how we developed subject-specific musculoskeletal models that account for the mechanical (e.g., muscle weakness) and motor control (e.g., spasticity) impairments affecting children with CP. Then, I will focus on the computational aspects and on the numerical methods we implemented to increase the computational efficiency of our simulations. Next, I will introduce how we aim to improve our understanding of gait control by using subject-specific predictive simulations. Finally, I will describe how we plan to combine those tools to create a simulation platform to predict gait performance following orthopedic interventions in children with CP.


30/03/2017(13:00) [Lieu : TFL seminar room]
Nicolas Bourgeois(iMMC)
"RANS simulations of Rapid Compression Machines using detailed chemistry: Impact of multi-dimensional effects on the auto-ignition of iso-octane and h-heptane"
Rapid Compression Machines (RCMs) have proven to be an effective experimental device for the investigation of auto-ignition phenomena. Comparing the experimental data gathered from RCMs with the simulated results obtained when using a specific kinetic mechanism is a key step in the development and the validation process of chemical kinetic mechanisms. In order to be consistent with the classic adiabatic core assumption used in such RCM numerical simulations, most of the pistons used in the experiments contain some crevices that have proven to make the reactive chamber homogeneous. The counterpart of using such a creviced piston is the mass transfer from the reaction chamber to the crevices induced during the preliminary heat release of two-stage ignition processes. This effect has been recently highlighted by Mittal and has been quantified to extend the auto-ignition delay up to 30%. Aiming to mitigate this mass transfer, the solution of ‘crevice containment’ (CC) has been proposed. It consists of a physical separation between the reaction chamber and the crevices region that is engaged at the end of the compression. In doing so, the benefits of using a piston with crevices during the compression stroke are maintained and the mass transfer to the crevices during the induction period is suppressed. In several studies, this novel CC concept has been experimentally and numerically (using a skeletal mechanism) assessed. It appears that the configuration with CC significantly mitigates the increase in auto-ignition delay induced by the crevice mass transfer. In the present study, reactive RANS simulations using detailed mechanisms are performed for geometries with and without the CC technology and are compared to the 0-D model for n-heptane and iso-octane. Due to the possible interactions between fluid mechanics and chemical kinetics, it is believed that high fidelity CFD simulations coupled to detailed chemistry can help gaining more insight on the phenomena arising within RCMs. This study has two objectives. First, it aims at quantifying the impact of the mass transfer to the crevices on the auto-ignition delay. Both n-heptane and iso-octane results highlight a behavior consistent with previous work: for two-stage ignition cases, the CC indeed meets the expectations and avoids the ignition delay extension due to crevices mass transfer. Second, the CC concept is evaluated for a larger temperature spectrum. Previous and current studies highlight the unavoidable formation of a vortex after the seal engagement due to the residual cylinder wall vorticity being prevented from going to the crevices. As a consequence, the ignition appears to be initiated in this ‘residual’ vortex region formed after compression for the temperature cases being inside the NTC region. The benefits of stopping the post- compression crevice mass transfer are thus counter-balanced by those unexpected effects that were not previously observed.


28/03/2017(14:30 - 17:15) [Lieu : EULER - Seminar room]
Prof. Dr. Rodney Fox(Iowa State University)
"Turbulence modeling for disperse multiphase flows. Application to fine-particle formation"


24/03/2017(13:00 - 13:55) [Lieu : EULER - Seminar room]
Sapanathan Thaneshan(iMMC / IMAP)
"In-situ formation of porous inner architecture at the magnetic pulse welded joints"
Nanoporous structures have multiple applications in various fields while their fabrication methods are very limited. Therefore, exploring the formation of porous structure will help to understand the underlying physics of the phenomena. This could lead to identify a potential method to produce the porous nature within a hybrid part or in bulk quantities. Porous structure formation resulted at the interface of electromagnetic pulse welded joints are investigated and revealed, to identify the development of sub-micron to few micron sized pores due to ultra-high rate of heating and cooling. Collision velocity along the overlapping length was predicted using a multi-physics numerical simulation and subsequently this velocity field was used to estimate the heating and cooling rates as about 109 °C s− 1. Predicted temperature, pressure and strain variations clearly evidence the observations of porous structure formation that is potentially facilitated by sequential phenomena of pores nucleation, coalescence and growth within the molten intermediate phase.


17/03/2017(16:45 - 17:00) [Lieu : Stévin - b. (-145)]
Sophie Heins(iMMC / MEED)
"Robotic-assisted serious game for motor and cognitive post-stroke rehabilitation"


07/03/2017(14:30 - 17:15) [Lieu : MERC 12]
Prof. Dr. Rodney Fox(Iowa State University)
"Mesoscale models for physical and chemical processes QBMM for spatially homogeneous flows"


28/02/2017(14:30 - 17:15) [Lieu : MERC 12]
Prof. Dr. Rodney Fox(Iowa State University)
"Quadrature-based moment methods (QBMM). The generalized population balance equation"


27/02/2017(14:00-15:00) [Lieu : Bât Euler - a.002 ]
Bishal Silwal(Aalto University)
"FEM in Electrical Machines and the Impact of Local Mesh Refinement on the Global Quantities"
Electrical motors account for about 45% of the global electric energy consumption in the world. On the other hand electrical generators are the major source of electrical energy. Therefore, there is a pressing demand on designing more energy-efficient rotating electrical machines. The development of computers has made the Finite Element Method (FEM) one of the most widely and commonly used tool for design and virtual prototyping. Rotating electrical machine models have however very specific features, compared for instance with FE models in other disciplines like Fluid dynamics or Structural mechanics. In the seminar, we talk about the simulation challenges in modeling rotating electrical machine, and take a closer look to the impact of local mesh refinement on the accuracy of relevant global quantities like torque and flux linkages. Three different simulation packages are used for these simulations: Comsol, Fcsmek and Gmsh/GetDP.


21/02/2017(14:30 - 17:15) [Lieu : MERC 12]
Rodney Prof. Dr. Fox(Iowa State University)
"Introduction to disperse multiphase flows. Mesoscale description of polydispers flows"
International Francqui Chaire 2016-2017 - Lecture Program - UCL Schedule


17/02/2017(16:15) [Lieu : Stévin, b. (-145)]
Thomas Mercier(iMMC / MEED)
"Variable-speed pumped hydro energy storage. Ancillary services capabilities and modelling for power systems studies"
Pumped hydro energy storage (PHES) is the most established technology for large-scale energy storage. While it is mainly used for balancing the grid on an hourly basis, with fixed power set points in pump and turbine modes, guide vanes control enables PHES units to provide frequency control in turbine mode. Recent developments in power electronics have provided PHES with a variable-speed feature, thereby extending the operation ranges and enabling new control strategies. For PHES units operated at a variable-speed, with these new control strategies, hydro turbine models traditionally used in power system studies have become irrelevant. In this presentation we will discuss the new operation ranges brought by the variable-speed feature, the possible control strategies as well as the frequency control capabilities. Modelling of variable-speed PHES for power system studies will also be covered and dynamic performances will be illustrated with simulation results.


10/02/2017(9:30-10:30) [Lieu : Bât Euler - a.002 ]
Jeanne Pellerin(MEMA)
"Decomposition of a hexahedron into tetrahedra"
In this seminar, I will focus on the following question: How many ways are there to subdivide a hexahedron in tetrahedra? Previous works suggest that 10 different patterns exist. I will show that there are most probably​ 3 times more and will describe the tools to achieve this enumeration.


03/02/2017(16:15) [Lieu : Stévin, b. (-145)]
Samuel Galle(UGent)
"Experimental optimization of an ankle-foot exoskeleton to reduce the metabolic cost of walking for practical applications in healthy and impaired subjects. "
Humans have never been able to accept the limitations of their bodies. One way to improve human performance in healthy subjects or restore performance for impaired subjects is the use of exoskeletons to assist locomotion. The metabolic energy use or the metabolic cost of exoskeleton walking is an important concern in exoskeleton research. Due to the complexity of the already highly efficient walking pattern, exoskeleton assistance needs to be optimal in order to reduce the metabolic cost of walking. My work focused on optimizing the human-exoskeleton interaction to reduce the metabolic cost of walking in order to use exoskeletons for practical applications in healthy and impaired subjects. To optimize the human-exoskeleton interaction, we first focused on optimizing the adaptation to exoskeleton walking and optimizing the exoskeleton assistance. These improvements allowed us to use the exoskeleton for practical applications in healthy subjects and in elderly. We showed that exoskeletons can improve the maximal walking performance and that our exoskeleton can reduce the metabolic cost of walking compared to unpowered exoskeleton walking in healthy elderly. We also performed a pilot test to show that patients with Chronic Obstructive Pulmonary Disease (COPD) can also walk with and benefit from an exoskeleton.


27/01/2017(11:00 - 12:00) [Lieu : UCL - Bât Euler - a.002 ]
Jean-François Remacle(MEMA)
"Monotone finite element scheme for harmonic equations in 2 and 3 dimensions"
Harmonic functions exhibit the mean-value property, which states that the average value of a function over a sphere is equal to the value of the function at the center point. An interesting consequence of this is that solutions of Laplace problems always reach their extrema’s at boundaries, and no local extrema can exist inside the domain. This is the maximum principle. It is of great importance in practice. When reparametrizing a triangulated surface, for instance, the maximum principle ensures the one-to-oneness of the mapping. Unfortunately, standard finite elements discretizations do not necessarily preserve the property at the discrete level. A standard finite element formulations based on the minimization of Dirichlet energy, is first presented, pointing out a geometrical relationship between angles in the mesh and the fulfillment of a discrete maximum principle. Then, a positive scheme is developed using the same geometrical relationship, yet using the mean value property. This second scheme is shown to be first order accurate for the L^2 norm of the error, but as accurate as the finite element scheme for the H^1 seminorm. We finally propose a scheme that is both positive and second order accurate.


16/12/2016(16:15) [Lieu : Stévin, b. (-145)]
Maxence Van Beneden(iMMC / MEED)
"Permanent magnet thrust bearing in ​flywheel energy storage system "


15/12/2016(13:00-14:00) [Lieu : Stevin, b.044]
Grosshans Holger(TFL)
"Is the Electrostatic Charging of Particles Predictable?”"


12/12/2016(14:00-15:00) [Lieu : UCL - Bât Euler - a.002 ]
Fengxiang Lin(MEMA)
"Deformation of Titanium alloys with body centered cubic structure: role of twinning "
Titanium (Ti) alloys are good structural materials for application in automobile and aerospace industries. Recently, several β-metastable Ti grades have been designed, which have both high strength and high ductility. Twinning plays an important role in the high performance of these Ti alloys. Crystal twinning occurs when two separate crystals share some of the same crystal lattice points in a symmetrical manner. In this work, a three-dimensional crystal plasticity finite element simulation is carried out to investigate the local stress fields around the deformation twins. Effects of the local stresses on the transmission of twinning through grain boundaries are also considered.


09/12/2016(16:15 - 17:15) [Lieu : Bât. Stévin - b. -145]
Joost Geeroms (VUB)
"The CYBERLEGs prosthesis: Investigating an energy-efficient active transfemoral prosthesis for different ambulation activities."
Lower limb amputees suffer from reduced mobility as a result of their limb loss. Most common prosthesis are passive devices, which are not able to provide a perfect energetic approximation of healthy limb for most ambulation tasks. The field of active prosthetics has improved over the past decade, yet there is still a big difference in weight with respect to classic prostheses and the improvement in terms of metabolic cost and gait symmetry is not evident. The CYBERLEGs prosthesis has been developed in two stages, finally resulting in an active device which enables amputees to perform sitting down/standing up tasks, step-over-step stair climbing and descent, slope walking and obstacle avoidance. For level walking, the developed compliant design allows for a reduced energetic consumption compared to other actuator types.


28/11/2016(14:00-15:00 ) [Lieu : UCL - Bât Euler - a.002]
Amaury Johnen(MEMA)
"Validation of finite element meshes: Bézier interpolation and a divide and conquer strategy to rule them all"
Curved high-order finite elements have gained interest over the recent years. Linear hexahedral meshes, on the other hand, have always been a must for several applications. Both kinds of meshes share however a common obstacle to their widespread use in mesh generators: their validation is a real challenge. In this seminar I will begin by introducing step by step the notion of finite element validity, the Bézier interpolation and its fundamental properties. Then, an efficient and elegant divide and conquer algorithm is presented that takes advantage of Bézier interpolation properties. Finally some possible extensions are investigated.


14/11/2016(14:00-15:00) [Lieu : UCL - Bât Euler - a.002 ]
François Henrotte(iMMC-MEMA)
"What Electromechanics teaches us about mechanical and electromagnetic laws"
We start by showing that electromagnetic (vector) fields and stress/strain tensor fields cannot be combined into a working electromechanical mathematical framework if the classical engineering formulations (which hide the role of metric, assumed Euclidean) are used. The more general concept of tensor field is then demonstrated step by step, on a simple geometrical basis (manifold, curve, vector, covector, tensor) and shown to eventually encompass all physical fields encountered in continuous medium theory, at the expense of having the metric tensor now explicitly apparent. The issues of differentiation (exterior derivative and Lie derivative) and discretization (shape functions) are then addressed in this more general context, and some important benefits for numerical simulation and optimisation (sensitivity) are pointed out.


24/10/2016(14:00-15:00) [Lieu : Bât Euler - a.002 ]
Alexandre Chemin(MEMA)
"Solving incompressible Navier-Stokes equations on GPU"
The method presented here is based on a spectral elements discretization and a high-order splitting scheme for incompressible Navier-Stokes equations. It allows solving half a billion dof problems on a single GPU.


10/10/2016(14:00) [Lieu : UCL - Bât Euler - a.002 ]
Jean-François Remacle(MEMA)
"Topological invariants in discrete meshes"
I will present here some discrete versions of the famous Hopf-Poincaré theorem for triangular meshes and quad meshes on general 2D manifolds.


07/10/2016(11h00) [Lieu : Auditorium SCES01]
Patrick Onck(Micromechanics, Zernike Institute for Advanced Materials, University of Groningen)
"Protein Mechanics: From amino-acid to swimming cells"
Proteins are often referred to as the building blocks of life, playing critical roles in almost all structures and activities in biology. At the smallest, atomic length scale, proteins consist of amino-acids that are covalently bonded together to form long chains. Depending on the amino-acid sequence the proteins fold into specific structures that enable their biological function. In this presentation, I will take you on a journey through the cell and discuss three different cellular functions that are dominated by protein mechanics, each at their specific time and length scale: (i) the beating of cilia and flagella that enable cells to swim, (ii) the mechanics of the cytoskeleton that endows the cell with structural integrity, and (iii) the highly-selective transport through nuclear nanopores that protects our DNA.


23/06/2016(11:00) [Lieu : Stévin, Room B044]
Gretar Tryggvason(University of Notre Dame (USA))
"Direct Numerical Simulations of Complex Multiphase Flows"
Direct numerical simulations (DNS), where every continuum length and time scale is fully resolved, allow us to follow the evolution of complex flows for sufficiently long time so that meaningful statistical quantities can be gathered. Results for relatively simple multifluid and multiphase systems with bubbles and drops in turbulent flows are now available, but new challenges are emerging. First of all, DNS of very large systems are yielding enormous amount of data that, in addition to providing physical insights, opens up new opportunities for the development of lower order models that describe the average or large-scale behavior. Recent results for bubbly flows and the application of statistical learning tools to extract closure models from the data suggest one possible strategy. Secondly, success with relatively simple systems calls for simulations of more complex problems. Multiphase flows often produce features such as thin films, filaments, and drops that are much smaller than the dominant flow scales and are often well-described by analytical or semi-analytical models. Recent efforts to combine semi-analytical models for thin films using classical thin film theory, and to compute mass transfer in high Schmidt number bubbly flows using boundary layer approximations, in combination with fully resolved numerical simulations of the rest of the flow, are described.


21/06/2016(11h00) [Lieu : Barb92]
Frederic Dubois(LMGC - CNRS - Université de Montpellier)
"Non Smooth Contact Dynamics, a relevant framework to model divided media"
Deriving a relevant numerical approach to model divided media raises many issues especially due to the fact that it mixes various time and space scales. In the present work we focus on a fully implicit approach, the Non Smooth Contact Dynamics method (NSCD), initiated by J.J. Moreau [1] and M. Jean [2], and implemented in the LMGC90 software [3]. The interaction laws, such as frictional contact, are written as multi mappings relating contact unknowns (impulse and relative velocity) without any additional regularization. In the frame of Convex Analysis proposed by J.J. Moreau a modified dynamics formulation is used, which is able to deal with non-smoothness; it is integrated with a first order scheme. Impulses account for all events supposed to occur during a time step: external loads, single or numerous impacts between pairs or agglomerates of contacting bodies. An iterative contact solver is necessary in order to exhibit the values of the unknowns. The method allows reasonably large time steps. Progressively the models simulated by non-smooth DEM become more complex:

Interactions between contacting bodies may be physically quite
complicated. Primary phenomena depicted by an interaction law are impenetrability of materials and dry friction. Other phenomena like rolling friction due to roughness, cohesive behavior, local plastic deformation, viscous friction or wear should be also taken into account.
Particle geometries become more complex introducing clusters of basic shapes, polyhedron or possibly non-convex shapes.
Multiple physics couplings: chemistry, electrical or thermal effects. Multi-phase mixture: gas-grain or fluid-grain media.
Complex grains behavior: elasticity, plasticity, fracture, etc.
Phase changing: solid/solids (fracture, sintering) or solid/fluid, etc.

As it will be detailed in this tals the Non Smooth Contact Dynamics method (NSCD) is a powerful modelling framework to derive numerical strategies. NSCD has been successfully used to study the behavior of granular materials and structures. It has been also used to study the behavior of masonry under seismic load, rock mass stability, fracture, multiple physics coupling, etc.

LMGC90 platform is a free collaborative software developed by several researchers, engineers and students: Michel Jean, Mathieu Renouf, Gilles Saussine, Alexandre Martin, R ?emy Mozul, Dominique Ambard, Cyril Bordreuil, Frdric Prales, etc.

[1] J.J. Moreau, Unilateral contact and dry friction in finite freedom dynamics, volume 302 of International Centre for Mechanical Sciences, Courses and Lectures. pp.182, J.J. Moreau, P.D. Panagiotopoulos, Springer, Vienna, (1988).
[2] M. Jean, The Non Smooth Contact Dynamics Method, in Computer Methods in Applied Mechanics and Engineering, special issue on computational modeling of contact and friction, J.A.C. Martins, A. Klarbring (ed), 177, pp. 235257, (1999).
[3] https://git-xen.lmgc.univ-montp2.fr/lmgc90/lmgc90_user/wikis/home


10/06/2016(16:15) [Lieu : Stévin, room b.044]
Patricia Leconte(iMMC/MCTR)
"Real-time performance-based smoothness assistance during rhythmic arm exercises "
Rhythmic and discrete movements are two fundamental motor primitives being - at least partially - controlled by different neural pathways. After a stroke, both primitives can be impaired. However, current upper-limb therapies - both conventional and robot-assisted - train mainly discrete functional movements. In order to recover the complete motor repertoire, training both movements should be offered with dedicated exercises. During the seminar I will elaborate a new performance-based robotic assistance to train rhythmic movements with a rehabilitation robot. This assistance has been validated on simulated an experimental data.


08/06/2016(14:00) [Lieu : Stevin building, room b.044]
Jean-Jacques Orban de Xivry (KULeuven)
"Humans might not be as effective as robots for reaching movements"
Movement planning consists of choosing the intended endpoint of the movement and selecting the motor program that will bring the effector on the endpoint. It is widely accepted that movement endpoint is updated on a trial-by-trial basis with respect to the observed errors and that the motor program for a given movement follows the rules of optimal feedback control. Here, we show clear limitations of these theories because of the existence of a switching cost for motor planning. First, this cost prevented participants from tuning their motor program appropriately for each individual trial. Second, we found that randomly changing the width of a target over the course of a reaching experiment prevents the motor system from updating the endpoint of movements on the basis of the performance on the previous trial if the width of the target has changed. These results provide new insights into the process of motor planning and how it relates to optimal control theory. It also highlights the limitations in using bio-inspired strategies for controlling robots. This talk takes place in the framework of a mini-symposium about "robotics and motor control": http://perso.uclouvain.be/renaud.ronsse/documents/symposium_VRG.pdf


08/06/2016(14:50) [Lieu : Stevin building, room b.044]
Wisama Khalil(Ecole Centrale de Nantes)
"eneral Dynamic solution for Floating Base Tree Structure Robots with Flexible Joints and Links"
This paper presents a general algorithm for solving the dynamics of tree structure robots with rigid and flexible links, active or passive joints, and with a fixed or floating base. The algorithm treats in a unified approach both the inverse and direct dynamics. It addresses also the hybrid case where each active joint is considered with known joint torque as in the direct dynamic case, or with known joint acceleration as in the inverse dynamic case. The main common feature of floating base, flexible joints, and links deformation degrees of freedom is that they are not actuated. Consequently, their accelerations cannot be prescribed, but rather have to be determined through the dynamic model. These three types of variables will be denoted as passive, and their accelerations must be determined in any dynamic problem. To that end, we propose a common framework based on the generalization of the efficient recursive Newton-Euler algorithms of Luh, Walker and Paul and of Featherstone, which are devoted respectively to the inverse and direct dynamic models of systems with rigid links and active joints. The link deformation is considered by the floating frame method, which is based on considering that the total motion of each flexible link is the result of the linear deformations of a mobile (and rigid) reference configuration to which is attached a rigid frame. The general algorithm is easy to program either numerically or using efficient customized symbolic techniques. It is of great interest for studying floating base systems with soft appendages as those currently investigated in soft bioinspired robotics or when a robotic system has to modify its structure for some particular tasks; such as transforming an active joint into a compliant flexible one, or modifying a task with a floating base into one with fixed base. This talk takes place in the framework of a mini-symposium about "robotics and motor control": http://perso.uclouvain.be/renaud.ronsse/documents/symposium_VRG.pdf


08/06/2016(16:00) [Lieu : BARB91]
Virginia Ruiz Garate (UCLouvain)
"Bio-Inspired motor primitives for controlling leg exoskeletons"
Locomotion assistive devices have gained increasing attention during the last years. In order to enhance the control of these devices, an emerging approach deals with capturing biological principles to emulate desirable features of human locomotion. In this thesis, a novel assistive controller based on the bio-inspired concept of motor primitives is presented. Motor primitives are seen as fundamental units of action, which through proper recombination can generate a high-dimensional set of stimulations. These stimulations activate the body muscles and thus generate movements. This work focuses on the particular use of primitives for assisting rhythmic locomotion tasks. Also, other bio-inspired locomotion mechanisms based on feedback are explored: short-loop reflexes and torso-balance control. In simulation, a bipedal walking model is developed by combining these different sources of muscular stimulation. This tool allows to explore the relative contributions of these mechanisms in human locomotion. Next, this thesis focuses on the validation of the controller for delivering assistance using artificial primitives during different experiments. The first one illustrates that the controller is able to work in real-time while providing assistance. The second set of experiments tests the controller performance when being used with a gait-impaired subject. Finally, the last set of tests aims at challenging the full capability of the controller by assisting healthy subjects during different locomotion tasks. These experiments highlight the capacity of volunteers to naturally interact with the device and generally benefit from the assistance. Importantly, when assisting different locomotion tasks, the controller is able to effectively handle the transitions between them without needing to stop in between. This event takes place in the framework of a mini-symposium about "robotics and motor control": http://perso.uclouvain.be/renaud.ronsse/documents/symposium_VRG.pdf