Jony Castagna

Science and Technology Facilities Council | STFC · Hartree Centre

Images and spectra obtained from aberration corrected scanning transmission electron microscopes (STEM) are now used routinely to quantify the morphology, structure, composition, chemistry, bonding, and optical/electronic properties of nanostructures, interfaces, and defects in many materials/biological systems. However, obtaining quantitative and...

Aerodynamic data required in aircraft high-value design and performance analysis are mainly acquired through wind tunnel testing and numerical analysis. Data acquisition can be expensive and is subject to multiple sources of uncertainty. Previously, the multifidelity data fusion framework introduced by Lam, Allaire and Willcox (2015) has been adapt...

We present a novel algorithm for the many-body Dissipative Particle Dynamics (DPD) forces calculation which allows to efficiently scale the DL_MESO software package on Multiple General Purpose Graphic Processing Units. Together with the extension to 64-bit integer arrays and addition of hard surface boundary conditions, the proposed algorithm allow...

View Video Presentation: https://doi.org/10.2514/6.2022-3526.vid Designing an aircraft and analysing its performance requires uncertainty-aware and robust aerodynamic data. The three principal ways of acquiring such aerodynamic data are flight testing, wind tunnel testing and numerical analysis. They all can be expensive and are subject to multiple...

Sub-sampling during image acquisition in scanning transmission electron microscopy (STEM) has been shown to provide a means to increase the overall speed of acquisition while at the same time providing an efficient means to control the dose, dose rate and dose overlap delivered to the sample. In this paper, we discuss specifically the parameters us...

Many-body electronic responses such as dispersion and polarization (at and beyond dipole order) present fundamental challenges in the simulation of materials at the molecular scale. To address these, an emerging strategy employing embedded quantum oscillators as a coarse-grained representation of such responses has been effective in predicting mate...

AI surrogate models for science, based on Deep Neural Networks (DNN), are becoming increasingly popular. This is especially the case for generative models such as Autoencoders and Generative Adversarial Networks (GANs), which are showing promising results and exciting opportunities in several fields of science. We present here an overview of the cu...

A multi-GPGPU development for Mesoscale Simulations using the Dissipative Particle Dynamics method is presented. This distributed GPU acceleration development is an extension of the DL_MESO package to MPI+CUDA in order to exploit the computational power of the latest NVIDIA cards on hybrid CPU–GPU architectures. Details about the extensively applic...

Direct numerical simulation (DNS) of a turbulent flow over an axisymmetric hill has been carried out to study the three-dimensional flow separation and reattachment that occur on the lee-side of the geometry. The flow Reynolds number is ReHReH = 6500, based on free-stream quantities and hill height (H ). A synthetic inflow boundary condition, combi...

A multi-block high-order finite-difference direct numerical simulation (DNS) code has been developed for studying turbulent flows over complex geometries. A global mapping technique is adopted and applied to a multi-block domain arrangement, in which each block is consisted of a structured mesh and the adjacent blocks are overlapped and matched wit...

Simulation of a turbulent °ow over an axisymmetric hill has been carried out to study three-dimensional °ow separation and reattachment that occur on the lee-side of the geome-try. A total of four simulations are considered at low Reynolds number of ReH=6500, based on free-stream quantities and hill height (H). Synthetic in°ow generation and data f...

The "slug capturing" method developed by Issa & co-workers is based on the numerical solution of the two-fluid model equations using fine meshes (unlike existing methods employing much coarser ones) to enable the resolution of inherent flow instabilities that would otherwise be averaged out. The method had been extensively validated in the past aga...

A high-order finite-differences direct numerical simulation code is developed for studying turbulent flows over complex geometries. The solver uses a global mapping based multi-block arrangement, with each block consisting of a structured mesh and the adjacent blocks overlapping each other along the interfaces. The physical bounds of interfaces are...

A direct numerical simulation of a round under-expanded sonic jet in turbulent supersonic cross flow has been carried out. The objective is to investigate the mixing mechanism which occurs downstream the jet by use of a passive scalar variable. The Navier-Stokes equations in the cylindrical geometry are solved by use of multiblock overlapping meshe...

Software re-engineering of a direct numerical simul ation (DNS) code for highperformance computation of turbulent flows has been carried out. The SBLI parallel highorder finite-difference code was primarily develope d for simulation of a shock wave interacting with a turbulent boundary layer developing over a bump geometry. The code has proved to b...

Software re-engineering of a direct numerical simulation (DNS) code for high-performance computation of turbulent flows has been carried out. The SBLI parallel high-order finite-difference code was primarily developed for simulation of a shock wave interacting with a turbulent boundary layer developing over a bump geometry. The code has proved to b...

An in-house DNS code has been developed over the years and validations have been carried out on various flow problems, including boundary-layer, plain channel, bump flow, and many other cases. This finite difference code solves full three-dimensional compressible Navier-Stokes equations using high-order (4th-order) for spatial derivatives and multi...