Search

With the increase in computational power in the last decade and the forthcoming Exascale supercomputers, a new horizon in computational modelling and simulation is envisioned in combustion science. Considering the multiscale and multiphysics characteristics of turbulent reacting flows, combustion simulations are considered as one of the most computationally demanding applications running on cutting-edge supercomputers. Exascale computing opens new frontiers for the simulation of combustion systems as more realistic conditions can be achieved with high-fidelity methods. However, an efficient use of these computing architectures requires methodologies that can exploit all levels of parallelism. The efficient utilization of the next generation of supercomputers needs to be considered from a global perspective, that is, involving physical modelling and numerical methods with methodologies based on High-Performance Computing (HPC) and hardware architectures. This review introduces recent developments in numerical methods for large-eddy simulations (LES) and direct-numerical simulations (DNS) to simulate combustion systems, with focus on the computational performance and algorithmic capabilities. Due to the broad scope, a first section is devoted to describe the fundamentals of turbulent combustion, which is followed by a general description of state-of-the-art computational strategies for solving these problems. These applications require advanced HPC approaches to exploit modern supercomputers, which is addressed in the third section. The increasing complexity of new computing architectures, with tightly coupled CPUs and GPUs, as well as high levels of parallelism, requires new parallel models and algorithms exposing the required level of concurrency. Advances in terms of dynamic load balancing, vectorization, GPU acceleration and mesh adaptation have permitted to achieve highly-efficient combustion simulations with data-driven methods in HPC environments. Therefore, dedicated sections covering the use of high-order methods for reacting flows, integration of detailed chemistry and two-phase flows are addressed. Final remarks and directions of future work are given at the end. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a gluing method for composite meshes. Different meshes are generated independently and are glued together using some extension elements to connect them. The resulting global mesh is non-conforming and consists of connected overlapping meshes. The method is inherently implicit, parallel and versatile, in the sense that it is PDE independent. The most cited gluing method is probably the Chimera method, used for overset grids, where patch meshes are superimposed onto a background mesh. The method employed here was originally devised for such situations and is now applied to disjoint or overlapping meshes. One of the advantages of the method is that the meshes do not have to coincide and can present a gap between them. The method is illustrated through some simple examples to demonstrate the mesh convergence and finally applied to the solution of the airflow in the complete respiratory system, by joining independent meshes for the large and small airways. [Copyright &y& Elsevier]

Abstract: This paper presents several parallelization aspects of Fluid-Structure Interaction (FSI) problems in computational mechanics when using an Arbitrary Lagrangian-Eulerian (ALE) scheme. The physical domain of the coupled problem is then solved on two different zones: a first zone for the fluid dynamics and the fluid mesh deformation and a second one for the solid mechanics. The idea can be further extended by adding more Physics to the coupled system, such as heat transport (for fluid and solid) or excitable media, among many others. In this paper, the basic two premises are that all problems can already be solved individually in parallel with good scalability and that the coupled system is solved in a coupled way within the same code. The paper introduces the formulation, presents some parallelization issues and proposes how to attack them, presents some results and discuss them and draws some future lines. [Copyright &y& Elsevier]

We present a new overlapping Dirichlet/Robin Domain Decomposition method. The method uses Dirichlet and Robin transmission conditions on the interfaces of an overlapping partitioning of the computational domain. We derive interface equations to study the convergence of the method and show its properties through four numerical examples. The mathematical framework is general and can be applied to derive overlapping versions of all the classical nonoverlapping methods. [Copyright &y& Elsevier]

We present a Chimera method for the numerical solution of incompressible flows past objects in relative motion. The Chimera method is implemented as an iteration-by-subdomain method based on Dirichlet/Neumann(Robin) coupling. The DD method we propose is not only geometric but also algorithmic, for the solution on each subdomain is obtained on separate processes and the exchange of information between the subdomains is carried out by a master code. This strategy is very flexible as it requires almost no modification to the original numerical code. Therefore, only the master code has to be adapted to the numerical codes and the strategies used on each subdomain. As a basic flow solver, we a use stabilized finite element method. [Copyright &y& Elsevier]

• The Integral Length Scale Approximation Subfilter Scale Model is applied to separated flows. • A procedure to determine the model constant is proposed. • A low-dissipation finite-element method is used. • The model accounts for interactions between numerical error and turbulence modelling errors. • Flows over a sphere and an Ahmed body are computed, and good agreement with the data is obtained. A new model for the unresolved stresses in large-eddy simulations was recently proposed by Piomelli et al. [J Fluid Mech 2015; 766:499–527] and Rouhi et al., [Phys Rev Fluids 2016; 1(4):0444011], in which the length scale is not related to the grid size, but determined based on turbulence properties. This model, the Integral Length-Scale Approximation (ILSA), has a single parameter, s τ , which represents the contribution of the unresolved scales to the momentum transport, and is assigned by the user. We test ILSA in complex geometries using a low-dissipation finite-element method, and propose a rational method to determine s τ on the basis of a grid-convergence study. The interaction of the model with the numerical method and grid topology is studied first; then, two cases are considered: the subcritical flow around a sphere, and the flow over the Ahmed body, a simplified car model. In each case calculations are performed using three grids and varying s τ. With a consistent combination of grid size and s τ the statistical results are in very good agreement with DNS data and experimental measurements. The eddy viscosity is insensitive to sudden variation of the mesh size, and the model adjusts to the different dissipation and diffusion characteristics associated with different grid topologies and numerical techniques. [ABSTRACT FROM AUTHOR]

The nasal valve is the narrowest part of the nasal airway which is responsible for the largest part of the nasal resistance. Even little changes in the aperture can affect the flow downstream through the nose significantly. Its principal function is to limit airflow for example during a rapid and short inhalation, also called a sniff. Coupling Computational Fluid Dynamics (CFD) with Fluid–Structure Interaction (FSI) allows solving and exchanging force and displacement between the solid and fluid domains and offers a more accurate representation of the physical system in confined flow cases. Furthermore, particle transport and deposition are performed in this study to reveal the effect of the complex coupling on the nasal cavity deposition of inhaled aerosols. Two different configurations are used to model the nasal valve and differences in magnitudes in deformations are observed during the sniff. A comparison between FSI results and the in-vivo evaluation of the deformation shows an acceptable agreement as to the first step of validation. In addition, the results demonstrated that FSI increases significantly the particle deposition in the nasal cavity and the micro-particle diameter is the critical range parameter to enhance deposition with nasal valve deformation during a sniff. • The results demonstrated that FSI increases significantly the particle deposition in the nasal cavity. • The micro-particle diameter is the critical range parameter to enhance deposition with nasal valve deformation during a sniff. • The Comparison between FSI results and the in-vivo evaluation of the deformation shows an acceptable agreement as first step of validation. [ABSTRACT FROM AUTHOR]

Because of their widely generalized use in many industries, composites are the subject of many research campaigns. More particularly, the development of both accurate and flexible numerical models able to capture their intrinsically multiscale modes of failure is still a challenge. The standard finite element method typically requires intensive remeshing to adequately capture the geometry of the cracks and high accuracy is thus often sacrificed in favor of scalability, and vice versa. In an effort to preserve both properties, we present here an extended finite element method (XFEM) for large scale composite fracture simulations. In this formulation, the standard FEM formulation is partially enriched by use of shifted Heaviside functions with special attention paid to the scalability of the scheme. This enrichment technique offers several benefits since the interpolation property of the standard shape function still holds at the nodes. Those benefits include (i) no extra boundary condition for the enrichment degree of freedom, and (ii) no need for transition/blending regions; both of which contribute to maintaining the scalability of the code. Two different cohesive zone models (CZM) are then adopted to capture the physics of the crack propagation mechanisms. At the intralaminar level, an extrinsic CZM embedded in the XFEM formulation is used. At the interlaminar level, an intrinsic CZM is adopted for predicting the failure. The overall framework is implemented in ALYA, a mechanics code specifically developed for large scale, massively parallel simulations of coupled multi-physics problems. The implementation of both intrinsic and extrinsic CZM models within the code is such that it conserves the extremely efficient scalability of ALYA while providing accurate physical simulations of computationally expensive phenomena. The strong scalability provided by the proposed implementation is demonstrated. The model is ultimately validated against a full experimental campaign of loading tests and X-ray tomography analyzes. [ABSTRACT FROM AUTHOR]

In view of the ongoing COVID-19 pandemic and its effects on global health, understanding and accurately modelling the propagation of human biological aerosols has become crucial. Worldwide, health professionals have been one of the most affected demographics, representing approximately 20% of all cases in Spain, 10% in Italy and 4% in China and US. Methods to contain and remove potentially infected aerosols during Aerosol Generating Procedures (AGPs) near source offer advantages in reducing the contamination of protective clothing and the surrounding theatre equipment and space. In this work we describe the application of computational fluid dynamics in assessing the performance of a prototype extraction hood as a means to contain a high speed aerosol jet. Whilst the particular prototype device is intended to be used during tracheotomies, which are increasingly common in the wake of COVID-19, the underlying physics can be adapted to design similar machines for other AGPs. Computational modelling aspect of this study was largely carried out by Barcelona Supercomputing Center using the high performance computational mechanics code Alya. Based on the high fidelity LES coupled with Lagrangian frameworks the results demonstrate high containment efficiency of generated particles is feasible with achievable air extraction rates. • Evaluation of a new device to reduce the risk of infection during Aerosol Generating Procedures for use in COVID-19 surgical tracheotomy. • High fidelity LES coupled with Lagrangian frameworks is used as results to demonstrate the efficiency. • Accurate numerical modelling of the propagation of human biological aerosols. [ABSTRACT FROM AUTHOR]

Accurate prediction of deformation and collapse of the upper airway during breathing is required for effective and personalised treatment of obstructive sleep apnoea (OSA). While numerical modelling techniques such as fluid–structure interaction (FSI) are promising, an outstanding challenge is to accurately predict the deformation of the airway during breathing and thus the occurrence of OSA. These difficulties arise because the effective stiffness of the soft tissue in the human upper airway varies due to neuromuscular effects on the stiffness of the underlying muscles. In addition, both the elasticity and anisotropy of the soft tissues along the upper airway are poorly characterised. Finally, gravitational effects on anatomic features are yet to be considered. In this study, a validated FSI technique is introduced that allows prediction of the extent and position of the major deformation in the upper airway. This technique is used to analyse the behaviour of the upper airway in the two most common sleeping positions and for a range of effective tissue stiffnesses. The results demonstrate that sleeping position, gravity and soft tissue stiffness (used here as a proxy for neuromuscular effects) are the main factors that affect upper airway collapse. Therefore, this study provides new insights into the mechanisms of OSA and a new methodology that significantly advances the patient-specific treatment of OSA. [ABSTRACT FROM AUTHOR]