This one day networking event on the 30th Oct at the University of Sheffield will bring together members of the N8 community who share a common interest in the topic of ‘Multiscale Computational Mechanics’ with a focus on structural continuum computational methods. The greatest challenges that we have to face are often of multiscale nature. The material behaviour must commonly be described at the microscopic, mesoscopic and macroscopic levels in order to understand the overall behaviour of the material.
Multiscale methods need to be developed across scales in order to design and manufacture safer, more reliable and more effective products. However, in order to tackle the great complexity associated to such multiscale approach, efficient computational approach must be developed and often High Performance Computing (HPC) systems must be harnessed to perform large calculations. Such complex multiscale nature is true for any discipline such as bioengineering, structural civil engineering, aerospace, automotive industry, material science, tribology, etc.
Therefore, this event will aim to draw academic and industrial researchers across disciplines together to mutually inform each other on the latest multiscale approach in their own field. We wish to promote inter-breeding across sectors and promote the establishment of new collaborations within the N8 HPC consortium as a result of the discussions within this event.
We hope to see longevity from the event with the ambition to develop it in to an annual networking day – informed by the best available experience and expertise.
Further information on topics to be covered during the day and information on key note speakers can be found in the event brochure.
Attendees are welcome to submit abstracts (max. 500 words) through the event registration form. The closing date for abstract submission is 5pm on the 20th of September 2015. Normal registration will remain open until 5pm on 19th October 2015.
Local organiser: Prof. Damien Lacroix, University of Sheffield
||Registration & Welcome Refreshments
||Prof Mike Hounslow
Pro-Vice-Chancellor, University of Sheffield
||Computational modelling of the mechanics of metallic biodegradable stents
||Prof Peter McHugh
National University of Ireland Galway
||A bottom-up approach to describe protein-protein interactions in a continuum mechanics biomolecular model
||Dr Albert Solernou
University of Leeds
||Morning Refreshment Break
||Multiscale modelling of the musculoskeletal system
||Dr Pinaki Battacharya
University of Sheffield
||Multi scale analysis of large composite structures
||Prof Silvestre Pinho
Imperial College London
||Civil engineering design using the Exascale desktop
||Dr Lee Margetts
University of Manchester
||Multiscale modelling of material failure
||Dr Karl Travis
University of Sheffield
||Microscopic modelling of fibre reinforced composite – a DEM approach
||Prof Jianqiao Ye
||Multiscale modelling of moisture diffusion coupled with stress distribution in CFRP composites
||Mr Maozhou Meng
||Afternoon Refreshment Break
||Debate: HPC opportunities in multiscale computational mechanics
|| Chaired by Professor Damien Lacroix
University of Sheffield
||N8 HPC Presentation
||Dr Robin Pinning
||Prof Damien Lacroix
University of Sheffield
Prof. Peter McHugh. Presentation title- “Computational Modelling of the Mechanics of Metallic Biodegradable Stents”. Guest Speakers:
Stents have revolutionised the treatment of arterial disease. Acting as a supporting scaffold, these small mesh devices are now routinely inserted into arteries where the blood flow has become dangerously restricted. In relation to coronary stents, one of the most fertile technological growth areas is biodegradable stents; here there is the possibility to generate stents that will break down in the body once the initial necessary scaffolding period is past (6-12 months) and when the artery has remodelled (including the formation of neo-intima). This brings advantages including the possibly of reduced risk of in-stent restenosis and late stent thrombosis, and the restoration of vasomotion potential. This is a very exciting technology and stents based on both metal and polymer platforms are emerging.
In this presentation a method to simulate the degradation of metal stents (magnesium alloy) is summarised, and implications for stent scaffolding performance presented. A stent angioplasty computational test-bed has been developed by the authors, based on the Abaqus software (DS-SIMULIA, USA), capable of simulating stent tracking, balloon expansion, recoil and in-vivo loading, in an atherosclerotic artery model. Additionally, a surface corrosion model has been developed and calibrated against experimental corrosion data for Magnesium alloy AZ31, and implemented in Abaqus/Explicit. This model takes two forms, a Uniform Corrosion model and a Pitting Corrosion model, and in both cases, material removal through corrosion is simulated by individual element removal from the finite element mesh. The model (in both forms) is implemented in the computation test-bed and used for stent analysis and design simulations. Recommendations are made on future stent design in relation to fundamental material properties and the optimization of stent geometry to maximize stent scaffolding support.
Prof. Silvestre Pinho. Presentation title- “Multi scale analysis of large composite structures”.
For the efficient structural design of large composite components, the numerical analysis of their mechanical response often requires different parts of the structure to be modelled at multiple length- and time-scales, eventually even using different physics. To this purpose, it is crucial to develop (i) suitable techniques for coupling areas of the structure discretized using different finite element types and (ii) numerical methods to efficiently compute equivalent homogenized properties to be used in both 2D FE models and in the coarse-scale subdomains of multiscale FE models of large composite components.
Regarding the first point, a novel Mesh Superposition Technique (MST) for the progressive transition between differently-discretized subdomains will be presented. The MST is applied to the multiple length/time-scale analysis of a low-velocity impact on a composite plate. Unlike using a sudden discretization-transition approach, the use of the MST avoids the undesirable stress disturbances and wave reflections at the interfaces between differently-discretized subdomains; therefore, it correctly captures the impact- induced damage pattern at a lower computational cost.
To address the second point, a novel set of PBCs named Multiscale Periodic Boundary Conditions (MPBCs), that represents the first set of PBCs that apply to reduced Unit cells (rUCs) and enable the direct two-scale (solid-to-shell) numerical homogenization of periodic structures, including their bending and twisting response, will be presented. Therefore, the MPBCs enable the use of rUCs to obtain the full ABD matrix needed in the coarse-scale subdomains of the MST models, with a significant computational cost reduction compared to the use of standard Unit Cells (UCs).
The results of these analyses demonstrate the relevance of the proposed approaches for an efficient multiple length/time-scale simulation of large composite structures.
Dr. Lee Margetts. Presentation title- “Civil engineering design using the Exascale desktop”.
One of the many challenges in multiscale modelling is dealing with increased computational complexity and computational cost for realistic simulations which involve different time and length scales. Exascale computing is expected to be with us in a couple of years and this new capability will be a game changer for multiscale modelling. But how long will it take these technologies to benefit industry? A recent survey from NAFEMS has shown that the majority of engineers in industry prefer to use their own workstation for engineering analysis, with little use of commercial software on HPC systems. This presentation will give an overview of the challenges of developing multiscale modelling software for Exascale systems. In particular, it will set out how open source academic software could form the basis of a new generation of modelling tools for the Exascale desktop of 2030. The talk will focus on examples of multiscale modelling in the field of civil engineering.