Alejandro Jacobo Cabrera Crespo

SPHysics - development of a free-surface fluid solver - Part 1: Theory and formulations

A free-surface fluid solver called SPHysics is presented. Part 1 provides a description of the governing equations based on Smoothed Particle Hydrodynamics (SPH) theory. The paper describes the formulations implemented in the code including the classical SPH formulation along with enhancements like density filtering, arbitrary Lagrange-Euler (ALE) schemes and the incorporation of Riemann solvers for particle-particle interactions. Second-order time stepping schemes are presented along with the boundary conditions employed which can handle floating objects to study fluid-structure interaction. In addition, the model implementation is briefly described. This information will be used in Part 2, where the efficiency of the code is discussed, along with several study cases.

GPUs, a New Tool of Acceleration in CFD: Efficiency and Reliability on Smoothed Particle Hydrodynamics Methods

Smoothed Particle Hydrodynamics (SPH) is a numerical method commonly used in Computational Fluid Dynamics (CFD) to simulate complex free-surface flows. Simulations with this mesh-free particle method far exceed the capacity of a single processor. In this paper, as part of a dual-functioning code for either central processing units (CPUs) or Graphics Processor Units (GPUs), a parallelisation using GPUs is presented. The GPU parallelisation technique uses the Compute Unified Device Architecture (CUDA) of nVidia devices. Simulations with more than one million particles on a single GPU card exhibit speedups of up to two orders of magnitude over using a single-core CPU. It is demonstrated that the code achieves different speedups with different CUDA-enabled GPUs. The numerical behaviour of the SPH code is validated with a standard benchmark test case of dam break flow impacting on an obstacle where good agreement with the experimental results is observed. Both the achieved speed-ups and the quantitative agreement with experiments suggest that CUDA-based GPU programming can be used in SPH methods with efficiency and reliability.

SPHysics - development of a free-surface fluid solver - Part 2: Efficiency and test cases

This paper, the second of a two-part series, analyses the efficiency of SPHysics and illustrates its capabilities by means of several test cases. Some intrinsic features of the SPH technique such as the use of link lists and the check for the limits are analysed here in detail. Numerical results are compared to experimental data for several cases studies: (i) Creation of waves by landslides, (ii) Dam-break propagation over wet beds and (iii) Wave-structure interaction. In addition, the capabilities of SPHysics to deal with realistic cases are depicted using the GPU version for several visual examples.

New multi-GPU implementation for smoothed particle hydrodynamics on heterogeneous clusters

Abstract A massively parallel SPH scheme using heterogeneous clusters of Central Processing Units (CPUs) and Graphics Processing Units (GPUs) has been developed. The new implementation originates from the single-GPU DualSPHysics code previously demonstrated to be powerful, stable and accurate. A combination of different parallel programming languages is combined to exploit not only one device (CPU or GPU) but also the combination of different machines. Communication among devices uses an improved Message Passing Interface (MPI) implementation which addresses some of the well-known drawbacks of MPI such as including a dynamic load balancing and overlapping data communications and computation tasks. The efficiency and scalability (strong and weak scaling) obtained with the new DualSPHysics code are analysed for different numbers of particles and different number of GPUs. Last, an application with more than 10 9 particles is presented to show the capability of the code to handle simulations that otherwise require large CPU clusters or supercomputers.

Optimization strategies for CPU and GPU implementations of a smoothed particle hydrodynamics method

Abstract Much of the current focus in high performance computing (HPC) for computational fluid dynamics (CFD) deals with grid based methods. However, parallel implementations for new meshfree particle methods such as Smoothed Particle Hydrodynamics (SPH) are less studied. In this work, we present optimizations for both central processing units (CPU) and graphics processing units (GPU) focused on a Lagrangian Smoothed Particle Hydrodynamics (SPH) method. In particular, the obtained performance and a comparison between the most efficient implementations for CPU and GPU are shown using the DualSPHysics code.

3D SPH Simulation of large waves mitigation with a dike

The interaction between a large wave and a coastal structure is studied with a three-dimensional (3D) Smoothed Particle Hydrodynamics (SPH) model. The role of protecting barriers (dikes and seawalls) to mitigate the force and moment exerted on the structure is analyzed in terms of the dike height and the distance from the dike to the structure. The existence of different propagation modes (different ways for the water to surpass the protection barrier) has been identified. In general, the flow is split into two parts: one overtopping the barrier and the other one flowing around it. The interaction between both the parts of the fluid is shown to be responsible of the force and moment exerted on the coastal structure.

SPH–DCDEM model for arbitrary geometries in free surface solid–fluid flows

Abstract A unified discretization of rigid solids and fluids is introduced, allowing for resolved simulations of fluid–solid phases within a meshless framework. The numerical solution, attained by Smoothed Particle Hydrodynamics (SPH) and a variation of Discrete Element Method (DEM), the Distributed Contact Discrete Element Method (DCDEM) discretization, is achieved by directly considering solid–solid and solid–fluid interactions. The novelty of the work is centred on the generalization of the coupling of the DEM and SPH methodologies for resolved simulations, allowing for state-of-the-art contact mechanics theories to be used in arbitrary geometries, while fluid to solid and vice versa momentum transfers are accurately described. The methods are introduced, analysed and discussed. Initial validations on the DCDEM and the fluid coupling are presented, drawing from test cases in the literature. An experimental campaign serves as a validation point for complex, large scale solid–fluid flows, where a set of blocks in several configurations is subjected to a dam-break wave. Blocks are tracked and positions are then compared between experimental data and the numerical solutions. A Particle Image Velocimetry (PIV) technique allows for the quantification of the flow field and direct comparison with numerical data. The results show that the model is accurate and is capable of treating highly complex interactions, such as transport of debris or hydrodynamic actions on structures, if relevant scales are reproduced.

SPHysics-FUNWAVE hybrid model for coastal wave propagation

It is difficult to study the process of wave propagation from the deep ocean to the nearshore region using a single model due to the presence of multiple scales both in time and in space. Numerical models based on the Boussinesq equations are well known to accurately propagate waves from intermediate water depth to the nearshore region. Since they are 2D models, they are computationally efficient and can be applied to study wave transformations over large domains. Numerical models based on Smoothed Particle Hydrodynamics can inherently capture multiply connected free surfaces and hence can be naturally used to capture breaking free surfaces and estimate breaking induced runup and overtopping. Here, a hybrid model (SPHunwave) is developed combining the main advantages of a Boussinesq model (FUNWAVE) and a SPH model (SPHysics). The details of the coupling procedure along with preliminary validation tests are presented.

Smoothed Particle Hydrodynamics for Water Waves

Smoothed Particle Hydrodynamics provides a numerical method particularly well suited to examine the breaking of water waves due to the ability of the method to cope with splash. The method is a meshfree Lagrangian method that allows the computational domain to deform with the flowing liquid. Here we discuss the appropriate kernels used in the interpolation and the time stepping alogrithms. Applications to water waves are shown. Copyright ?? 2007 by ASME.

Integration of UAV photogrammetry and SPH modelling of fluids to study runoff on real terrains.

Roads can experience runoff problems due to the intense rain discharge associated to severe storms. Two advanced tools are combined to analyse the interaction of complex water flows with real terrains. UAV (Unmanned Aerial Vehicle) photogrammetry is employed to obtain accurate topographic information on small areas, typically on the order of a few hectares. The Smoothed Particle Hydrodynamics (SPH) technique is applied by means of the DualSPHysics model to compute the trajectory of the water flow during extreme rain events. The use of engineering solutions to palliate flood events is also analysed. The study case simulates how the collected water can flow into a close road and how precautionary measures can be effective to drain water under extreme conditions. The amount of water arriving at the road is calculated under different protection scenarios and the efficiency of a ditch is observed to decrease when sedimentation reduces its depth.