Programma
Le informazioni sul programma sono in aggiornamento continuo. Si prega di visitare spesso la pagina per gli ultimi aggiornamenti o modifiche.
MERCOLEDI' 7 SETTEMBRE 2022
19:30 Cocktail di benvenuto
GIOVEDI' 8 SETTEMBRE 2022
10:15-11:45 Coffee break - Visita a Palazzo Bo
19.30 Visita all'Università di Padova - Orto Botanico
20.00 Cena sociale all'Orto Botanico dell'Università di Padova
VENERDI' 9 SETTEMBRE 2022
16.00 Chiusura dei lavori
Plenary lectures
MULTISCALE ANALYSIS OF VERY LARGE STRUCTURES
Modern composite structures, such as an aircraft wingbox, are typically very complex structures 10s of meters long. A single-scale analysis of an aircraft winbox with the level of detail required to represent damage propagation within composite plies is computationally unfeasible. In fact, even successfully modelling a complete aircraft wingbox with the level of detail required to reliably identify damage hotspots is computationally challenging; for instance, a non-linear wingbox simulation, capable of supporting the prediction of failure phenomena, will involve 100's of interacting parts and may exceed 100MDoF in size. Solution of a single configuration scenario will generate a results database >100Gb in size, containing billions of individual failure indices and damage variables --- manual analysis of data on such a scale is intractable. In this work, we present a redesign of the traditional failure modelling approaches specifically aimed at very large engineering structures. Specifically, we present (i) a harmonized modelling framework across various scales of analysis, as the same code is accessed via different subroutines that are used for different levels of idealization; (ii) a methodology for the calculation of high-value data directly in the computer cluster during the analysis, so that the output file eventually accessed by the user can be used effectively for decision making, such as identifying hot-spots for a multiscale analysis. We also show ongoing work whereby this high-value data is directly used by the large-scale analysis itself to identify hot-spots and adapt the mesh and idealization concurrently with the simulation. Finally, we show how the framework described can support underpinning fundamental research with a clear route for industrial application, and how it provides an effective route for design of very large composites structures using physically-based and effective analysis methods.
Silvestre Pinho Imperial College London (UK). Silvestre Taveira Pinho is a Professor in Mechanics of Composites in the Department of Aeronautics at Imperial College London. He was awarded in 2010 by the European Society for Composite Materials (ESCM) the prize for best young researcher in Composites active in Europe. In 2014, Silvestre was awarded an EPSRC fellowship for designing novel forms of more damage tolerant composite structures. Silvestre is a member of the Council and of the Executive Committee of the European Society for Composite Materials (ESCM); of the Executive Council of the International Community for Composite Materials (ICCM); and of the Royal Aeronautical Society Structures & Materials Specialist Committee. Silvestre's main research focus is on experimental, analytical and numerical aspects of failure in fibre-reinforced composite materials and structures. Other interests include the design of improved microstructures, recycling and reusing composites for a sustainable future and the design of graphene-based composites. Silvestre’s main research contributions include proposing translaminar fracture toughness of fibre-reinforced composite plies as a material property, as well as a corresponding test method for measuring it. This enabled a type of regularised numerical models for failure of composites that is now mainstream. The team led by Silvestre in the second World-Wide Failure Exercise ranked top in terms of its blind failure predictions. Some of Silvestre’s failure models for composites currently ship natively in both Abaqus and LS-Dyna. Silvestre has proposed several bio-inspired microstructures for composites, some of which are currently being further developed for actual industrial applications.
UNRAVELING THE FATIGUE BEHAVIOUR OF METAL LATTICE STRUCTURES PRODUCED BY LASER POWDER BED FUSION
Laser Powder Bed Fusion (LPBF) is a widely used Additive Manufacturing (AM) technique which allows production of complex shaped solid and porous samples and components in metal. Starting from a CAD file, parts are generated layer by layer by melting of powder particles in successive tracks and layers. Very complicated porous lattice structures can be designed and produced with LPBF. The mechanical properties of these structures are depending on the parent material, the lattice density, the unit cell’s geometrical characteristics and the presence of possible surface and volume defects. All these factors are also related to LPBF specific process and post-process parameters. Quasi-static behaviour of LPBF produced metal lattices is fairly well understood, while fatigue behaviour is more complex and less documented. This presentation aims to partially address this, by summarizing a decade of research on fatigue of metal lattices produces by LPBF, and by isolating and discussing the various fatigue influencing factors as well as tools and measures to extend, calculate and predict the fatigue life of metal lattices produced by LPBF.
Brecht Van HoorewederKU Leuven (Belgium). BRECHT VAN HOOREWEDER obtained his PhD in 2013 at the Mechanical Engineering Department of KU Leuven in Belgium. He is currently leading the Additive Manufacturing team and the Additive Manufacturing Institute at KU Leuven as tenured professor and as successor of Prof. Jean-Pierre Kruth. He is supervising 4 postdocs & 17 PhD students and he is coaching a multidisciplinary team working on additive manufacturing of metals, ceramics and polymers. The expertise and research activities of Brecht his team is currently mainly focused on development and in-line real-time process monitoring & control of novel laser based additive manufacturing processes and on studying the fundamental fourfold machine-process-microstructure-properties relationships in solid and porous metals produced by such new additive manufacturing techniques. Brecht Van Hooreweder is also coordinating one of the best equipped academic laser based AM labs in the world, benefitting from 3 decades of AM research by Prof. Kruth, and comprising three in-house developed Laser Powder Bed Fusion (LPBF) machines that have unique features such as microwave preheating, closed loop melt pool control, pulsed laser erosion, etc. combined with 7 commercially available printers, dedicated equipment for powder handling and mechanical testing of AM parts, and state-of-the-art safety procedures.