Code Developers’ Workshop 2

Bio: Professor Xue-Feng Yuan is a Chartered Physicist, a Fellow of the Institute of Physics and a Fellow of Royal Society Chemistry, UK. He has joined Guangzhou University as Professor in Systems Rheology since November 2016 after three year service as the Founding Director of National Supercomputer Centre in Guangzhou for running “Tianhe-2 (Milky Way-II)” supercomputer which was ranked as the world fastest supercomputer six times successively from June 2013 to June 2016. He is a Vice-President of Chinese Society of Rheology and the China Delegate of the International Committee of Rheology (ICR). He also serves several international organisations, including the BRICS Innovation Collaboration Working Group on ICT and HPC, the Asia-Pacific Regional Innovation Knowledge Network for 4th Industrial Revolution Technologies and the Global Initiative Working Group of the European Commission eInfraCentral Coordination Action. Professor Yuan has over 30 years research experience in experimental, theoretical and computational rheology of soft condensed matter, recently its applications in biology and medicine. He obtained PhD from the Victoria University of Manchester, UK in 1989. Then he worked as a postdoctoral research associate at Theory of Condensed Matter Centre, Cavendish Laboratory, the University of Cambridge, UK for five years, and worked as a Kao Corporation visiting scientist and as a part-time lecturer in Department of Applied Physics, Nagoya University, Japan. He won a prestigious 5-year EPSRC Advanced Research Fellowship in 1996 and worked in Centre of Theoretical Physics, Department of Physics, the University of Bristol, UK till 1999, then joined Department of Mechanical Engineering, King’s College London, UK. He took a Readership in Biochemical Physics at Manchester Interdisciplinary Biocentre, School of Chemical Engineering and Analytical Science, the University of Manchester, UK, in 2004. He served various EPSRC Prioritisation Panels for Processing Engineering, Structural Materials, Physics, Fighting against Crime Programme, and MRC Prioritisation Panel for Discipline Hopping Grants. He was an external PhD thesis examiner for the University of Cambridge, Warwick University, University of Sheffield and Swansea University.

Bio: Dr. John-Paul Latham has a PhD in Structural Geology and is Reader in Geomechanics at Imperial College. He has championed the use of discontinuum solid numerical models based on the combined finite-discrete element method (FEMDEM or FDEM). The modelling platform which was started by Latham and Munjiza’s £0.81M EPSRC project, ‘VGW’ (Virtual Geoscience Workbench, 2004-2009), was re-launched in 2016 as Solidity with contact mechanics and fracture modelling showing unprecedented accuracy. He was PI on: eight EPSRC grants; itf-ISF3 petroleum industry consortium on fractured reservoirs; NERC UK’s HYDROFRAME developing models for hydro-thermo-mechanical and fracturing processes in fractured rocks around a geological disposal nuclear waste facility; H2020 project ‘SURE’ on radial jet drilling for Geothermal Resources, and H2020 project ‘ORCHYD’ on hybrid water jet-percussion drilling. His extensive knowledge of rock for coastal protection led to him writing the 240-page Chapter 3 on materials, for the ‘The Rock Manual. The use of rock in hydraulic engineering’ C683 CIRIA/CUR/CETMEF, 2007, and his FDEM modelling research on breakwater stability began in 2002. He has >200 publications (94 Journal) and has given keynotes in diverse fields including “Components in an understanding of rock blasting” and “How do rubble breakwaters survive wave attack? Challenges for a FEMDEM/CFD model solution”. He is lead supervisor on 16 PhD projects.

Bio: Dr Jiansheng Xiang is a research fellow in Computational Mechanics. Jiansheng has a great passion for research in applied mathematics and modelling and has continuously diversified his efforts to numerous fields of engineering mathematics and its application areas: computational fluid/solid mechanics, fluid-solid interactions, heat transfer, fracture mechanics, and turbulent flow. He has published over 50 peer-reviewed articles in leading engineering and computational physics journals in addition to 76 conference papers. These cover many areas in engineering mathematics including Combined Finite-Discrete Element Method, Discrete Element Method, fracture models, nonlinear finite elements, coupled multi-scale modelling, immersed body methods, reduced order modelling, and mesh adaptivity, and a very diverse range of applications in Catalysis engineering, Coastal engineering, Reservoir geomechanics, Bio-engineering, rock blasting, pneumatic conveying, and decarbonisation technology including tidal energy and hydrogen production using steam methane reforming.

Dr Jiansheng Xiang is the principal developer of the solid modelling code, Solidity (previously named VGeST), which is a suite of Open Source Tools for modelling discontinuous, particulate, granular, blocky, fracturing and fragmenting systems. It is the first 3D multibody transient dynamic model of its type (FEMDEM) and was uploaded on (https://sourceforge.net/projects/vgw) on 18/09/2008. Since its launch, it has achieved >6,000 downloads across 125 countries in the world. He has also developed a novel and efficient approach for fluid-structure interaction: the immersed body method for fluid-solid coupling by combining Solidity solids models with AMCG’s world-leading adaptive mesh Computational Fluid Dynamics (CFD) model, Fluidity.

Bio: Dr. Philip Cardiff holds a post as Bekaert Lecturer of Materials Processing. A graduate of Mechanical Engineering, Philip completed a PhD at UCD in the area of computational biomechanics. Upon completing his PhD, Philip spent time as a Post Doctoral Research Fellow working on the development of computational models within the Irish Centre for Composites Research at UCD, and within the Cockrell School of Engineering at the University of Texas at Austin.His research interests include Computational Mechanics, Materials Processing and Biomechanics.

Abstract: SWIFT is a particle hydrodynamics code aimed (currently) at simulating the evolution of the Universe and the formation of galaxies. The code achieves an order of magnitude speed-up compared to the “industry standard” code gadget. Adaptive spatial sampling and time stepping is essential for any cosmological or particle code, but this, combined with the complex physics, makes conventional approaches to parallelism challenging. SWIFT is the result of a radical rethink that places fine-grained task parallelism at the heart of the code design. I will describe some of the lessons learned and our plans to take this forward in the next generation of the code.

Bio: In the pre-pandemic world, Professor Richard Bower would be found drinking coffee at the Institute of Computational Cosmology at the University of Durham. He has a track record of papers that have helped shape our understanding of the formation of galaxies and the processes that shape the observable Universe. His current interest is in developing better numerical simulation codes and in applying emulation techniques to better understand the macroscopic processes that emerge from the detailed local physics.

Abstract: Big promises are being made for how quantum computing will transform computational capabilities in the near future. The promises are based on solid theoretical foundations, but realising the benefits will need hard work and creativity to develop useful quantum algorithms. Initially at least, these will run as subroutines to augment classical codes, hence quantum hardware must also interface effectively with classical HPC. Finding the bottlenecks where a quantum subroutine could significantly speed up computations is also a good excuse to revisit and revitalise old algorithms: the best quantum method does not always directly correspond to the best known classical method for the same problem. There are several initiatives in the UK focused on enabling the computational science and engineering communities to develop quantum enhancements to their codes. CCP-QC was funded in the 2019 CCP renewal call to bring existing CCPs together with quantum computing experts, to identify early use cases, develop testbed implementations, and train early career researchers in quantum computing. The UK National Quantum Computing Centre is being established on the STFC Harwell campus, to provide a bridge between scientific prototypes and commercial quantum hardware and software. Among other roles, it will house testbed hardware, enabling codes to be tested in readiness for deployment when quantum hardware of sufficient power is available.

Bio: Dr Viv Kendon joined Durham University Physics Department in August 2014. She is a member of the Quantum Light and Matter (QLM) Research Section and the Joint Quantum Centre (JQC) Durham-Newcastle. She held an EPSRC established career fellowship on Hybrid Quantum and Classical Computation from 2014-2019. Prior to Durham, she was part of the Quantum Information Group at the University of Leeds, where she held a Royal Society University Research Fellowship from 2004 to 2012. Dr Kendon was educated at the Universities of Oxford and Edinburgh, and initially researched in the area of Soft Condensed Matter, using latticeBoltzmann methods to simulate fluid mixtures. She switched to quantum information theory via postdoctoral positions held at the University of Strathclyde and Imperial College. Previous to her research career, Dr Kendon was active for over 10 years in global electronic networking and computer support in the voluntary sector. Her current main research interests are the computational physics of quantum information theory, and interfacing quantum and classical computational architectures. She is chair of CCP-QC.

Abstract: The numerical modelling can represent a useful and complementary tool to physical model tests. Sophisticated tools are now at a formative stage and here we are actively developing the novel numerical technique Smoothed Particle Hydrodynamics (SPH). As a meshless and Lagrangian technique, SPH is ideally suited to fluid and solid mechanics with highly nonlinear deformation. SPH describes a fluid by replacing its continuum properties with locally smoothed quantities at discrete Lagrangian locations. Thus, the domain can be multiply-connected with no special treatment of the free surface, making it ideal for examining complicated flow situations. The DualSPHysics code has been developed to use SPH for real engineering problems. Among the different SPH codes, the DualSPHysics software is considered one of the most efficient SPH solvers. DualSPHysics is open-source and allows applying the SPH method to real life engineering problems. It can be executed not only on CPUs, but also on GPU (Graphics Processing Unit) cards with powerful parallel computing. The DualSPHysics code has been coupled with the Project Chrono library to reproduce the complex mechanisms, such as the ones involved in PTO systems, within the same meshless framework. DualSPHysics has proven its ability to generate and propagate waves, and to satisfactorily simulate wave interactions with wave energy converters (WEC). Mooring lines can be also handled by the library MoorDyn that has been also coupled with DualSPHysics. DualSPHysics code can be proposed now as a complementary tool to physical model experiments for a preliminary design of WEC devices.

Bio: Alex Crespo is Associate Professor at the University of Vigo. He obtained his PhD in Applied Physics in 2008 at the University of Vigo. He has 16 years of experience in research and teaching. His research activity is mainly focused on computational fluid dynamics and its application to coastal engineering and wave energy devices. In particular, he has been developing open-source codes based on the Smoothed Particle Hydrodynamics (SPH) method. He is also a member of the Steering Committee of SPHERIC, which is the international organisation that aims to promote the development and the use of SPH in academia and industry.

Abstract: This talk will focus on a general-purpose and high-performance scientific code coupling library called the Multiscale Universal Interface (MUI). This library is released as open-source and is publicly available, representing a collaborative effort between UKRI-STFC, Brown University, Lawrence-Berkeley National Laboratory and IBM Research UK. The goal of MUI is not to prescribe how codes should be coupled, rather it provides an extensible toolkit to enable the underlying mechanisms required for various types of partitioned coupled problems. The talk will present the library in terms of its design and capabilities and its fundamental performance on HPC systems, offering an insight into where it has already been applied and its scope for future development..

Bio: Stephen is a Principal Computational Scientist within UKRI-STFC’s Scientific Computing Department. He has a background in computer science, including a PhD that saw computer science topics considered within the context of physical modelling at the University of Manchester, which in turn led to a domain specialism in computational engineering. His research activities vary within this remit but he has a particular focus on particle-based simulation techniques such as Smoothed Particle Hydrodynamics and, more recently, general scientific code coupling and its applications. Stephen is involved in the UK HPC community, with all of his work underpinned by this general thread and a particular interest in exploiting upcoming heterogeneous architectures.

Bio: Mikito Furuichi is a Group Leader of Computational Science and Engineering Group in the Center for Mathematical Science and Advanced Technology at the Japan Agency for Marine-Earth Science and Technology, and is also Visiting Associate Professor, Kobe University. He researches Computational Geodynamics and the Numerical Schemes associated with that. He develops the parallel algorithms of particle simulations, robust Stokes flow solver, and efficient treatment of numerical boundaries for HPC, for solving the geodynamics, civil, and granular engineering problems. His algorism enabled numerical experiments using 2.4 billion DEM particles to understand the 3D stress state in the accretionary prism.

Bio: Prof Hrvoje Jasak is one of the original Co-Authors of OpenFOAM and the Gianna Angelopoulos Lecturer in Scientific Computing in the Department of Physics at the Cavendish Laboratory of the University of Cambridge. He is the Director of Wikki Ltd (CFD and OpenFOAM consultancy and development), Professor at the University of Zagreb, Croatia and Mercator Fellow at TU Darmstadt, Germany. Prof Jasak gained his first degree in mechanical engineering from the University of Zagreb, Croatia in 1992 and his PhD in CFD from Imperial College London in Prof. Gosman’s group (1993-1996). His research interests include: Numerical simulation methods (FVM and FEM), specifically Computational Fluid Dynamics; Object-oriented design, expert C++ programmer, Unix/Linux, high-performance computing; Numerical modelling of free surface flows, naval hydrodynamics and wave modelling; Mathematical modelling of continuum phenomena and numerical mathematics (linear solver technology); Dynamic mesh handling, error estimation, adaptive mesh refinement; and, Modelling of complex heat and mass transfer systems.

Bio: Alessandro Franci is Assistant Research Professor at International Center for Numerical Methods in Engineering (CIMNE) and at the Universitat Politècnica de Catalunya (UPC). He obtained the Master Degree in Civil Engineering at the Politecnico di Milano in 2011 and the PhD degree in Structural Analysis at the UPC in 2015. The PhD thesis of Franci was selected ex aequo as the best PhD thesis in numerical methods in Spain for the year 2015 by the Spanish Society of Numerical Methods in Engineering (SEMNI). His main expertise is on the development of particle-based computational methods for engineering problems. He is also the director of the PARTICLES COURSE, an ECCOMAS advanced course aimed at giving an overview of theory and applications of the most used particle methods. The research lines of Franci include the numerical simulation of geophysical flows, free-surface fluids, fluid-structure interaction, thermal-coupled problems, and particle-laden flows.

Bio: Gerasimos Chourdakis is a doctoral candidate at the Department of Informatics, Technical University of Munich. A chemical engineer from NTUA (Athens), who continued with a M.Sc. in Computational Science & Engineering at TUM. He is part of the preCICE team since 2017 and contributes, among other packages, to the OpenFOAM-preCICE adapter. His ongoing PhD research focuses on geometric multi-scale coupling, with nuclear reactor simulations as a main use case, while his interests also extend to (free) research software engineering. Read more at https://www5.in.tum.de/~chourdak/.

Bio: Benjamin Uekermann is a junior professor within the Cluster of Excellence SimTech at the University of Stuttgart. He has an interdisciplinary background in between computer science, applied mathematics, and mechanical engineering. His research focuses on the development of numerical methods and algorithms for multi-physics, multi-scale simulations, their efficient implementation on parallel systems, and their realization in easy-to-use, sustainable, and free software. Since 2012, Benjamin has been one of the main developers of the coupling library preCICE. Read more at https://github.com/uekerman..

Abstract: Due to the non-linear, time-dependent nature of various multi-physical problems, a multi-physical simulation tool that is numerically robust, highly scalable and general in purpose is challenging to develop. We start building an effective and robust multi-physical simulation tool from the framework on fluid-structure interaction (FSI) simulations. A partitioned approach is used, ensuring a separation of concerns (fluid, structure and coupling) and also allowing design flexibility and robustness while reducing future maintenance efforts. The Multiscale Universal Interface (MUI) general code coupling library is employed as the interface between fluid and structural domains. The computational fluid dynamics (CFD) package OpenFOAM and the computational structure mechanics (CSM) solver FEniCS are utilised for the fluid and structural domains respectively. Three explicit/implicit coupling utilities for the FSI coupling have been developed in the MUI library to achieve a tight and stable coupling. Validation and demonstration cases for both single-phase and multiphase FSI problems will be shown.

Bio:Dr Wendi Liu is a Senior Computational Scientist at the Daresbury Laboratory, working within the Engineering and Environment group of the Scientific Computing Department. Wendi’s research interests include Novel approaches to the simulation of fluid-structure interaction (FSI), Numerical analysis of thermo-elastic problems, Multi-physics code-coupling within a High Performance Computing (HPC) framework and FSI for Renewable energy devices (e.g. wind turbines).

Bio:Prof. Kevin Maki is the Richard B. Couch Development Professor of Marine Hydrodynamics, and the Director of the Aaron Friedman Marine Hydrodynamics Lab in the Department of Naval Architecture and Marine Engineering at the University of Michigan. His primary research area is marine computational mechanics with an emphasis on developing new methods for performance evaluation of marine vehicles. He is also active in developing new methods for the aerodynamic design of more fuel efficient ground vehicles.

Bio: Rajeev K. Jaiman is currently an Associate Professor and Seaspan Chair in the Department of Mechanical Engineering at the University of British Columbia (UBC), Vancouver, Canada. An aeronautical engineer by training, his research concentrates on a high-fidelity physical modeling and data-driven computing, with emphasis on large-scale computations of fluid-solid and fluid-fluid interface problems. Prior to his current appointment at UBC, he was an assistant professor in the Department of Mechanical Engineering at the National University of Singapore (NUS). Before joining NUS, Dr. Jaiman was the Director of Computational Fluid Dynamics (CFD) Development at Altair Engineering, Inc., Mountain View, California. The CFD technologies that Dr. Jaiman has developed are being used in wind turbine, offshore oil/gas, nuclear reactors, automotive and aerospace industries. Dr. Jaiman earned his Bachelor of Technology degree in Aerospace Engineering from the Indian Institute of Technology, Mumbai. Dr. Jaiman received his master’s and doctorate degrees in Aerospace Engineering from the University of Illinois at Urbana-Champaign (UIUC). In 2018, Dr. Jaiman was awarded the Institute of Engineers Singapore (IES) prestigious engineering recognition for the development of a next generation revolutionary semi-submersible, and in 2005 he was awarded by the Strehlow Memorial “Outstanding Researcher” recognition. He has published over 150 peer-reviewed journal and conference proceedings and has served as expert reviewers for numerous leading journals. He is a senior member of AIAA and members of ASME, USACM, APS and SIAM

Abstract: Observations from previous tsunami events often reveal variation in the damage and survivability of impacted similar structures. Such variation can be due to shielding effects and other interactions that occur when the tsunami wave occurs in urban locations. In this talk, we will present the validation and application of the meshless numerical technique Smoothed Particle Hydrodynamics (SPH) to tsunami-structure interaction. As a meshless and Lagrangian technique, SPH is ideally suited to fluid and solid mechanics with highly nonlinear deformation such as the violent tsunami bore impact on the structures. We use the open-source DualSPHysics, jointly developed by the University of Manchester to explore the influence of onshore structures’ orientations and arrangements during a tsunami impact event. The results reveal significant reductions in total force on a structure can be made via strategic spatial positioning and orientation. Such reductions may mean the difference between superficial damage and complete structural collapse. This enables the development of more resilient structures in tsunami-prone regions.

Bio: Professor Benedict Rogers is a Professor of Computational Hydrodynamics at the University of Manchester, specialising in Smoothed Particle Hydrodynamics (SPH). He obtained his doctorate at the University of Oxford on the application of Riemann solvers to shallow flows, making some important fundamental developments. Thereafter he developed his interest in SPH as a Research Fellow at Johns Hopkins University. He leads the SPH@Manchester research group where he coordinates a team of 9+ PhD students, 3-4 PDRAs, visiting academics collaborating with leading international companies including NNL, Unilever, motorsport teams. He is co-founder of the SPH rEsearch and engineeRing International Community (SPHERIC), the international organisation representing the development of SPH and is leading the organisation as the current Chair. He is a core developer of the community open-source SPH codes SPHysics and DualSPHysics http://www.dual.sphysics.org downloaded over 77,000+ times with annual international users’ workshops. He has published over 65 peer-reviewed journal papers and was twice awarded the Thomas Telford Premium Award by the Institution of Civil Engineers (ICE) in 2014 and 2016 for his work on SPH..