Sessions

When available, a synopsis appears after mouse-clicking on the session title.

S1

Mechanics of polymers and metallic glasses - Experiments and models

Both polymers and metals can be in an organized crystalline or amorphous glassy state, where for polymers usually at least a part of the structure is amorphous and metals are in a glassy state only when processed under special conditions. The session focuses on the mechanical properties of amorphous or partly amorphous solid materials and attempts to bridge descriptions found for metallic glasses and polymers in a glassy or rubbery state, including semi-crystalline polymers, which share some common features (such as a rate- and temperature-dependent response, being prone to strain localization in the form of shear bands, the occurrence of damage by cavitation etc.). Contributions and new developments in the characterization of their elastic-(visco)-plastic response and early stages of damage and failure at different length scales are of interest, as well as continuum and/or atomistic modelling approaches for their mechanical behaviour, including fracture. Combined experimental and numerical approaches for proper parameter identification and predictions are also welcome.

The topics of this session include, but are not limited to:
  • Influence of processing conditions and production methods
  • Time and temperature dependent properties
  • Constitutive modelling
  • Creep and failure
  • Ab-initio, atomistic, continuum and multi-scale modelling methods
  • Experimental characterization methods
  • Novel applications

 

Rafael Estevez    Hans van Dommelen
(Grenoble INP, France)     (TU Eindhoven, The Netherlands)

S2

Mechanics of composites - Experiments and models

Composite materials exhibit outstanding mechanical properties which are ideal for their use in lightweight driven structural applications. However, the lack of reliable and experimentally contrasted models to predict accurately the failure strength and damage evolution makes the optimization process of composite structures to rely on costly and time-consuming experimental campaigns. The difficulties in the simulation of mechanical behavior of composites are endorsed to the natural competition between failure mechanisms operating at different scales from micro to the macro level. Model identification and validation with advanced experimental techniques as well as recent constitutive developments for novel structural composites (textiles, advanced fibers, etc.) will be also welcome. The session includes, but is not limited to the following topics:
  • Micro and meso mechanics of structural composites
  • Advanced models for novel composite architectures
  • Multiscale methods to bridge length scales
  • Constitutive modeling of deformation and fracture mechanisms.
  • Virtual testing. Effect of defects.
  • Experimental mechanics applied to modeling and parameter identification (e.g. digital image correlation, in-situ experiments and tomography, etc.)

 

Soraia Pimenta     Carlos Gonzalez
(Imperial College, United Kingdom)      (IMDEA Madrid, Spain)

S3

Mechanics of metals - Experiments and models

In order to establish predictability of mechanical behaviour of metallic materials, insights from the smallest length-scales are needed, along with the development of effective approaches to utilize such fundamental information at larger scales. In this session, we aim to bring experimentalists and modellers together to discuss these critical issues in light of new developments in the area of mechanics of metals. Issues relating to the spatial and temporal evolution of dislocation structures, the influence of microstructural content (e.g., grain boundaries, point defects, second phases) on the deformation, damage and failure mechanisms, and the interplay between experimental observations and predictive models at the nano- and micron-scales are central themes. We strongly encourage contributions that directly couple experiment and modelling approaches.

Topics of interest include:
  • microstructural characterization of deformation structures
  • dislocation-meditated deformation in small, finite or confined volumes
  • high fidelity observations or simulations of dislocation processes
  • size effects in mechanical behaviour
  • atomistic simulations
  • discrete dislocation dynamics
  • in situ testing (SEM, Synchotron, low or elevated temperature)
  • micromechanical constitutive laws (e.g., strain gradient plasticity)
  • multiscale bridging approaches

 

Ron Peerlings      Erica Lilleoden
(TUE Eindhoven, The Netherlands)    (HZ Geesthacht, Germany)

S4

Mechanics of ceramics - Experiments and models

Advanced and traditional ceramic materials represent an important research topic, connected to several industrial sectors operating in mechanical, aerospace, and nuclear engineering. In this field, continuous innovation and enhancement in the technology is imperative to maintain industrial leadership. The aim of this session is to gather together scientists working in the field of ceramic science and material simulations to present new approaches to mechanical modelling, particularly with reference to ceramic-matrix composites, ultra-high temperature ceramics or refractories, ceramics for biomedical applications, and for high efficiency energy technologies.

Topics of particular interest, but not exclusive, could be:
  • innovative testing and characterization of the mechanical behavior,
  • continuum modelling from brittle to ductile deformation behavior,
  • mechanical behavior coupled with microstructural change,
  • and fracture mechanics

 

Davide Bigoni      Nicolas Schmitt
(Università di Trento, Italia)   (LMT Cachan, France)

S5

Additive Manufacturing

Additive manufacturing is a growing technology allowing to build parts and complexes structures layer by layer. The technology has started on polymers and with non-critical parts but moved recently to metal and ceramic technologies. These shifts allowed considering that technology for more critical parts in demanding applications. In particular, the possibility to build optimized structures lead to re-thinking the use of material, a playground for topological optimization and links between structural topology and mechanical properties. In addition, the shift towards more demanding structural parts lead to the requirements of higher mechanical performances for these materials. The session will thus cover all aspects of the mechanics of materials related to additive manufacturing from process optimization (including post-processing) towards mechanical properties improvement trough structure optimization adapted to the constraints of the AM technologies, involving both experimental and numerical studies.

 

Aude Simar      Rémy Dendievel
(UCLouvain, Belgium)   (SIMAP Grenoble, France)

S6

Ductile damage and fracture

Ductile damage may be defined as the cause of failure processes involving a significant amount of dissipation which can occur at various length scales. Progression of ductile damage will ultimately lead to fracture and typically includes extensive deformation in the material non-linear regime. Damage constitutive models can be mechanism based or purely phenomenological in nature. Central to the progress made and the application of damage constitutive models in practice is the development of robust computational models and reliable experimental techniques for model calibration.

TOPICS of interest are envisioned in the following areas (but not restricted to):
  • Damage constitutive models,
  • Initiation of damage,
  • Model for critical damage,
  • Fracture at low stress triaxiality,
  • Experimental methods involving in particular in situ testing,
  • Numerical modeling.

 

Thomas Pardoen      Kim Lau Nielsen
(UCL Louvain, Belgium)   (DTU Lyngby, Danmark)

S7

Fatigue, reliability and lifetime predictions

Fatigue is a complex phenomenon that is influenced by many factors including microstructure, defects, surface state, environment, prior processing and loading. Understanding the nucleation and propagation of fatigue cracks in materials and structural  components subjected to cyclic loading is crucial for optimized design from material microstructures to structural engineering components. The identification and mechanistic basis of the physical processes leading to crack nucleation and growth across the length scales (from micro to macro) are still major issues in many engineering fields and the establishment of quantitative prediction remains a key research challenge. This session aims to bring together scientists and engineers  addressing fatigue research utilising experimental, characterisation and modelling approaches (ideally integrated) over many material systems.

The session covers the following areas, although the list is not restrictive:
  • Novel experimental methods to characterize fatigue damage and crack growth
  • Innovative theoretical approaches, computational and analytical methods
  • Experimental and numerical design and validation methods
  • Fatigue mechanisms in advanced alloys and metallic systems
  • Process - microstructure - fatigue interactions
  • Microstructure computation
  • Meta-materials and lattice structures
  • Fatigue in harsh environments (corrosion, low temperature, manufacturing processing etc.)
  • Overloads, arbitrary loading sequences, service spectrum loads, combined fatigue regimes
  • Residual stress effects on fatigue damage and crack growth, measurement of internal stresses
  • Thermal and thermo-mechanical fatigue
  • Microstructurally short cracks
  • Multiaxial fatigue life prediction
  • Fatigue crack propagation under complex (including non-proportional) loadings
  • Probability-based design criteria
  • Life prediction methodologies for structural materials
  • Damage tolerance and fatigue life
  • Fatigue of composites, polymers and elastomers
  • Fatigue of Shape Memory Alloys

 

Nicolas Saintier      Fionn Dunne    
(ENSAM Bordeaux, France)      (IC London, United Kingdom)     

S8

Failure of quasi-brittle materials

This thematic session is focused on the modelling of quasi-brittle material failure, understood as the progressive loss of the structural strength due to the formation of localization bands, propagating cracks and shear bands. The aim of the session is to provide a forum to present and discuss the progress, from theoretical, experimental and numerical point of view, in modelling of quasi-brittle materials, in particular concrete, rock and composites.

While contributions in all aspects of quasi-brittle material failure are invited, some of the topics to be featured are:
  • Continuous and discontinuous fracture modelling techniques
  • Local and nonlocal damage models
  • Phase-field and thick-level-set models
  • Dynamic fracture modelling
  • Multi-scale and multi-physics modelling techniques
  • Modelling of quasi-brittle metamaterials
  • Modelling of interface failure
  • Micro-mechanical modelling of quasi-brittle material failure
  • Models for proper representation of anisotropy

 

Claudia Comi      Bert Sluys
(Politcenico di Milano, Italy)    (TU Delft, The Netherlands)

S9

Functional and architectured materials

The investigation of functional and architectured materials has rapidly advanced in the last decade. For example, the combination of different phases within micro-heterogeneous composites, with individual physical characteristics of the individual constituents, allows for the design of materials with outstanding effective properties. The multi-physics nature of functional materials poses multiple challenges in scale-bridging, modelling, simulation and designing architectured materials with new functionalities. Recent advances in material manufacturing, such as 3D printing and additive manufacturing, open an exciting opportunity to realize and test pre-designed materials with desirable and pre-defined functionalities. This session aims to bring together experts in modelling, simulation, manufacturing and experimental investigation in the field of functional and architectured materials to present and discuss recent advances.

Topics of particular interest include (but not limited to):
  • 3D printing of architecture materials and composites
  • Electro- and magneto-active materials
  • Mechanical and Acoustic metamaterials
  • Responsive gels; Shape-memory and light-sensitive materials
  • Microstructural and material instabilities
  • Bio-inspired and biological materials
  • Ferroic and multiferroic materials
  • Selective Laser Melting (SLM) processes

 

Stephan Rudykh      Jörg Schröder
(Technion, Israel)   (UDE Duisbourg/Essen, Germany )

S10

Coupled problems in material mechanics

A large number of material mechanics problems require a multiphysics approach to correctly describe the material response at the length scale of interest. At a continuum level, this involves the formulation of complex multi-field constitutive models as well as efficient numerical procedures for solving coupled Boundary Value Problems. In addition, model identification based on experimental measurements or simulation results at a lower scale is often challenging. The aim of this session is to gather scientists working across a variety of material systems to share recent advances in the modelling and simulation of coupled problems in mechanics. Experimental approaches in direct relation with model identification and validation are also welcome.

Topics of interest include:
  • Thermo-mechanical, electro-mechanical, or magneto-mechanical couplings;
  • Mass transport and chemical reactions in solids;
  • Multiphysics phenomena in soft active materials, including elastomers, gels and biological tissues;
  • Biomechanical couplings;
  • Coupled-field problems in heterogeneous media;
  • Scale bridging approaches;
  • Computational approaches for coupled, multi-field problems.

 

Laurence Brassart      Andreas Menzel
(MU Monash, Australia )   (TU Dortmund, Germany)

S11

Mechanics of biological materials & biomechanics

Mechanics plays prominent roles in the study of biological systems and processes. It may advance our understanding of physiological and pathological mechanisms of organs, interaction between medical devices and biological material, drug delivery pathways, the interplay between structure and function of tissues, mechanotransduction and many others. Although to some extent traditional applied material mechanics concepts are directly applicable to describe biological materials, their inherent property to adapt to mechanical and chemical environments, remains a challenging modeling task. Likewise, in order to investigate biological organs adequately, i.e., to gain a comprehensive view of a biological process, sophisticated and robust material mechanical models are needed that couple among structural, fluid, chemical and other fields. Current material models are rather simplistic and much more concerning the load transition within biological tissues remains to be explored. Here, it would be crucial to understand not only the (passive) interaction of structural components within the tissue, but also how cells dynamically maintain such a structure. Biological tissues sense and responds actively to alterations in its mechanical environment, a fundamental tissue property that is currently largely overlooked in modelling biological tissues. For this session, we solicit contributions that address challenges directly related to modeling the mechanical properties of biological materials. This includes investigations at the organ, tissue and cellular levels. Contributions that consider
  • non-linearities and/or multiscale approaches
  • physiological and pathological mechanisms
  • active/growth/remodeling properties of biological tissues
  • clotting and thrombus formation modeling
  • uncertainty and sensitivity analyses
  • inverse and in-vivo parameter estimation
  • medical image-based studies are particularly welcome.

 

Vikram Deshpande      Christian Gasser
(UC Cambridge, United Kingdom)   (KTH Stockholm, Sweden)

S12

Mechanics of interfaces and evolving microstructures (incl. phase transformation and recrystallization)

The session proposes to gather contributions on the topic of evolving microstructures, with a particular focus on recrystallization, twinning and solid state phase transformations in metals and alloys. A better understanding of these microstructure evolutions implies the use of predictive models at different scales, which describe the complex evolutions of interface patterns, looking at the local kinetic equations, and at the global meso- or macroscopic resulting properties. These models include the so-called mean field models taking advantage of (partial) differential equations operating on well chosen state variables. They also refer to more demanding mesoscale discrete and continuum models and corresponding computational methods with explicit representations of microstructures through grids or meshes (Monte Carlo, Cellular Automata, Phase field, Level set, FEM, etc.). At the lowest scale, atomistic simulations provide new insights into the mechanisms operating during interface motion. Experimental approaches also explore the motion of microstructure interfaces at different scales, looking at crystallographic relationships between different grains or phases, dislocations structures induced by mechanical mismatch, interface mobility and energy, coupling between phase transformation – recrystallization - precipitation phenomena. In situ experiments at Large Facilities provide more and more information on those subjects, which need to be translated into appropriate mechanical and physical descriptions. At the laboratory scale, the possibility to explore microstructure evolutions in macroscopic samples from the measurement of temperature, stress/strain, strain rate, dilatometry, resistivity should be further investigated, in particular by taking advantage of multiscale models. The session welcomes all contributions related to the above topics, with possible extensions to polymers and composites, and to liquid-solid transformations, as long as the main focus remains on the description of moving interfaces.

 

Paul Steinmann      Roland Logé
(LTM Erlangen, Germany)   (EPFL Lausanne, Swiss)

S13

Mechanics of discrete systems (incl. granular media, tribology, scratch, indentation, adhesion)

Significant advances in contact mechanics has been achieved since the first theoretical derivations a few centuries ago. These are primarily associated with contact problems in statics and dynamics involving friction, adhesion, wear, roughness, heat or electric conduction, for various material behaviors. Principles of contacts mechanics can be applied to many traditional mechanical engineering areas such as locomotive wheel-rail contact, coupling devices, braking systems, tires, bearings, combustion engines, mechanical linkages, gasket seals, metal forming, ultrasonic welding, electrical contacts, and many others. Current challenges regard the extension of contact mechanics methodologies across scales, including the micro- and the nano-scale, to coupled multi-field problems, and the application to biological systems characterized by finite elasticity. The main objective of this thematic session is to gather specialists on theoretical, numerical and experimental modelling of contact problems in nature and technology, with special attention to innovative interdisciplinary research. Industrial applications of contact mechanics and the presentation of case studies of technological relevance are invited too.

 

Jean-François Molinari      Marco Paggi
(EPFL Lausanne, Swiss)   (IMT Lucca, Italia)

S14

Experimental mechanics (incl. advanced full-field deformation measurements and parameter identification)

The reliable simulation of complex material behaviour is driven by the desire to obtain a good prediction of the real process. A practical example of engineering interest is the predictive simulation of crash scenarios in the automotive industry. The simulation of manufacturing processes is another example, where it is preferable to use material models coupled to microstructure evolution during the process. Typically the development of constitutive equations is based on theoretical considerations, but experimental data are necessary to determine the constants, which enter these equations. Additionally, the models have to be investigated with respect to prediction. This step on validation again requires a broad basis of experimental data. The present minisymposium focusses on parameter identification as well as procedures for validation of constitituve models on different lenght scales. Presentations related to fluid mechanics, dynamics, geotechnics, multiscale methods, acoustics and electromagnetics are also welcome. In this way, the symposium is meant to bring together scientists from different research areas such as
  • Experimental investigations
  • Probabilistics
  • Mathematical optimization
  • Sensitivity analysis
  • Error estimation and adaptivity with finite elements
  • Model adaptivity
  • Industrial applications

 

Johan Hoefnagels      Rolf Mahnken
(TUE Eindhoven, The Netherlands)      (LMT Paderborn, Germany)

S15

Experimental nanomechanics

Nano- and micromechanical testing coupled with advanced in situ techniques has attracted significant attention over the past decade because it permits (i) probing individual microstructure constituents and (ii) selectively activate mechanisms. Both is key to fundamentally understand plasticity and fracture of advanced multi-phase microstructures and small scale systems. The main objective of this thematic session is to gather specialists in the field of Nano- and Micromechanical testing to unravel mechanisms of plasticity and fracture at interfaces.

Characterization techniques envisioned comprise (but are not limited to) transmission electron microscopy (TEM), scanning electron microscopy (SEM) and synchrotron based techniques. The contributions should aim for understanding plasticity, fracture and fatigue at
  • grain and phase boundaries
  • amorphous / crystalline interfaces
  • free surfaces
  • multilayered systems
  • super lattices.

 

Christoph Kirchlechner      Frédéric Mompiou     
(MPIE Düsseldorf, Germany)   (CEMES, Toulouse, France)  

S16

The mechanics of highly porous materials: experiments and modelling

Highly porous materials, also termed "cellular materials" contain a solid part that can be made of metal, polymer, ceramic or eventually composites and a gaseous part with a very high volume fraction. This encompasses wools, foams, lattices, textiles, preforms, ... This class of materials is omnipresent in nature (trabecular bone, wood, cork, coral, etc...) and is now also spreading as engineering material in the industry and as a topic of increasingly multidisciplinary research in the mechanics of materials community. Additive manufacturing is a very recent and attractive way of obtaining such structures with designed shapes. Highly porous materials have interesting physical properties (acoustic, thermal, ...), but before using them in structural applications, their mechanical properties have to be properly characterized and understood. In terms of mechanics, they are pressure sensitive, prone to size effects and can sometimes be better regarded as structures or systems rather than materials. The present symposium aims at gathering scientists interested in the mechanics of highly porous materials. Both experimental and theoretical/numerical approaches are welcome in the sessions.

 

Patrick Onck      Eric Maire
(Rijksuniversiteit Groningen, The Netherlands)   (INSA Lyon, France) 

S17

Coupled experimental-numerical approaches in materials characterization

To understand and predict the mechanical behavior of metallic materials, an in-depth understanding of the dominant mechanisms operating at different length scales is needed. This involves going from the atomic scale of dislocation cores to the nanometer and micrometer scales characteristic of the microstructure, all together responsible for the macroscopic mechanical behavior. Such widespread knowledge can only be gathered through a constant dialogue between experiments and simulations; dialogue, which has lately been promoted thanks to the development of advanced mechanical models and new experimental settings. In this session, the interplay between experimental observations and predictive models at the nano- and micron-scales is a central theme. We encourage contributions that directly couple experimental and numerical approaches at the same or different length scale.

Topics of interest include, but are not limited to:
  • Micro and Meso deformation experiments and simulations
  • Dislocation mechanisms and their observations
  • Fracture mechanisms and related experiments
  • Micromechanical constitutive laws and their validation against experiments
  • Modeling and experimental observations of phase transformations

 

Helena Van Swygenhoven      David Rodney
(EPFL Lausanne, Swiss)   (UL Lyon, France)

S18

Advanced modelling techniques: higher-order continua

Generalized continua are now extensively used to model and simulate size effects in the deformation and fracture of materials. Such approaches include strain gradient theories for elasticity, plasticity and damage, but also higher order media that introduce kinematic or physical degrees of freedom in addition to the usual displacement field. Cosserat and micromorphic approaches can be used to address a large variety of material behavior like foams, single and polycrystalline metals, composites, and soils and other civil engineering materials. Generalized material models can be identified from experimental data but also, increasingly, from micromechanical considerations based on extended homogenization theory. Various techniques are available like multiscale expansion methods or non homogeneous boundary conditions applied to representative volume elements. The session is open to any new advances in this field from the theoretical, computational and experimental perspectives.

Sessions are envisioned in the following areas, although the list is not exclusive:
  • Strain gradient plasticity theory and applications
  • Cosserat and micromorphic approaches
  • Gradient damage models
  • Non local theory and applications
  • Regularization methods for strain localization phenomena
  • Mesh-independent finite element simulations of fracture
  • Higher order approaches in geomechanics
  • Homogenization theory and generalized continua
  • Micromechanics of generalized continua
  • Dislocations and generalized continua
  • Grain size effects in metal polycrystals

 

Samuel Forest     Swantje Bargmann
(Mines ParisTech, France)   (HZ Geesthacht, Germany)

S19

Advanced modelling techniques: scale bridging

Multi-scale techniques and scale bridging play a key role in connecting the macroscopic behaviour of materials and structures at the engineering level directly to the material microstructure and microscopic deformation processes. Many different classes of scale bridging methods have been developed for this purpose. These generally involve multiple disciplines, e.g., engineering, computational mechanics, mathematics, physics, chemistry, and so on.

This session is intended as a forum for bringing together scientists from different disciplines working on multi-scale techniques and scale-bridging problems in mechanics of materials, including both spatial, as well as temporal scales.

The topics to be addressed in this session include (but are not limited to):
  • homogenization-based methods, e.g. mathematical homogenization, computational homogenization etc.
  • embedded domain methods, domain decomposition methods, global-local techniques
  • heterogeneous multi-scale method (HMM), equation-free method
  • (non-equilibrium) thermodynamics based coarse graining methods
  • methods for bridging distinct models, e.g. atomistic-to-continuum, quasi-continuum
  • methods for phenomena with (partially) non-separating scales, e.g. localization, damage and fracture or transient phenomena
  • methods for coupled multi-field phenomena (e.g. thermo-chemo-electro-mechanical etc.)
  • model reduction techniques and reduction of computational costs associated with multiscale algorithms and complex microstructures, e.g. arising from experimental imaging techniques
  • methods for atomistic to continuum scale-bridging

 

Varvara Kouznetsova     Bob Svendsen
(TU Eindhoven, The Netherlands)   (MPIE Dusseldorf, Germany)

S20

Advanced modelling techniques: Stochastics in materials mechanics

This session focuses on new theoretical, numerical and experimental approaches in the description of stochastic multi-scale microstructures and the prediction of their mechanical properties and behavior. In terms of microstructure description, the session covers in particular the topics of morphological analysis and segmentation of 3D microstructure images, stochastic models of disordered structures. Concerning the prediction of mechanical properties, the session covers all aspects related to stochastic homogenization in a broad way, including full-fields, finite element and Fourier-based numerical homogenization of random models and of real 3D microstructures, as well as analytical, numerical and experimental studies of scale-effects for random microstructures. Some applications include analytical and numerical probabilistic models for failure, fatigue and damage, and the morphological study of local fields. Finally, the session covers computational methods and algorithms for the simulation of heterogeneous and stochastic materials, including:
  • coupling schemes for stochastic multiscale simulations,
  • methods for wave propagation in random media,
  • and error estimation methods.

 

Régis Cottereau      Francois Willot      Carsten Proppe
(Ecole Centrale Paris, France)   (Mines ParisTech, France)   (ITM Karlsruhe, Germany)

S21

Advanced modelling techniques: phase field and diffuse-interface approaches

The phase-field method has over the years become the method of choice for simulating microstructure evolution at the micro- to mesoscopic scale. Applications reach from solidification over solid-state transformations including elasticity and plastic relaxation to piezo-electric and magnetic coupling, and the study of twinning and dislocation movement. The symposium shall review actual developments in this rapidly growing field with special emphasis on the interactions between the microstructure and mechanical loading.
  • Multi-scale approaches including or based on a phase-field model are of interest as well.
  • Materials in focus are metals and ceramics;
however applications to soft matter and polymers are welcome to explore synergetic approaches.

 

Benoît Appolaire      Nele Moelans
(ONERA, France)   (KU Leuven, Belgium)

S22   

Reduced order and data-driven material models

Although computation power constantly increases and numerical algorithms are improving quickly, many problems in materials science remain intractable today. A demonstrative example are multiscale approaches which are natural for the description of many physical processes which span many orders of scale in time and/or space. Recent advances in computational techniques now offer remarkable insights into materials behaviors at small scales where direct continuum-based descriptions might be insufficient. Therefore, to gain the benefits from the detailed insight into the material for application at larger scales, mathematical models must be nested and the information of the more detailed scale has to be coarse grained to carry the relevant information to the higher levels. However, this transfer is often unfeasible from computational point of view, in particular, if the problem at hand is large and non-linear. Model order reduction approaches, such as Reduced Basis approaches, Proper Orthogonal Decomposition or Proper Generalized Decomposition methods have proven to obtain approximate solutions in a reasonable computational time. By systematically reducing the model dimensions, the computational complexity can be reduced dramatically making the solution of problem at hand traceable. However, the identification of reduced-order models of large-scale systems is by far non-trivial. Here, data-driven approaches can significantly contribute where the model is designed by learning the system response characteristics from previously generated data. Especially the use of machine learning based tools, such as artificial neural networks or Gaussian processes, have recently demonstrated to be particularly successful for the identification of the key dependencies. Combining data-driven approaches and reduced order models lead to data-driven reduced models. The required data might be obtained from simulations at higher resolution or lower scale but might come also from experimental tests. Nowadays, large sets of experimental data are available, however, its incorporation into computational model is complex, requiring the development of algorithms and data handling techniques. Therefore, this session aims to bring together scientists from experimental characterization, constitutive modeling, model order reduction and machine learning to explore recent developments in these fields. Therefore, we invite contributions within these fields to explore the synergies of the different disciplines.

 

Benjamin Klusemann   Fransisco Chinesta
(LHZ Geesthacht, Germany)   (Ecole Centrale Nantes, France)
Online user: 1