Electromagnetic waves - Research Group of Dr. Conor Brennan

1. Electromagnetic wave scattering from complicated bodies

2. Envelope simulation of large scale circuits and systems

Astrophysics - Research Group of Dr. Turlough Downes

1. Propagation of stellar jets.

2. Turbulence in molecular clouds

3. Decay of magnetic fields around neutron stars (with Dr Sarah Tanner (PDRA)

4. Multifluid stability (with Dr Mohsen Shadmehri (PDRA) 

5. Stellar jet launching (funded under SFI RFP, hiring not complete yet)

Atomic, Molecular and Optical Processes  - Research Group of Dr. Lampros Nikolopoulos

Energy and Environmental Flow Modelling Group (Dr Yann Delaure, MME)

 Electromagnetic waves and circuits (Research Group of Dr. Conor Brennan)

1. Electromagnetic wave scattering from complicated bodies

The DCU RF modelling and simulation group are engaged in several strands of research which is compatible with high performance computing.  In particular we are using the integral equation formulation to solve the problem of electromagnetic wave scattering from complicated bodies.

The novel methods we are investigating originated in our work on modelling propagation over terrain and are now being applied to a wider range of problems including the modelling of microstrip patch antennas as part of our 2006 SFI PI project. The computational burden arises as the discretised integral equation takes the form of a large dense linear system which must be solved, often over a wide range of frequencies.

Our SFI RFP05 project involves developing efficient iterative matrix solvers that we have shown perform well when compared with the much-used Krylov solvers.  Complementary work using acceleration methods based on the fast multipole method to speed up the matrix vector multiplications within the solver will also be investigated.

A related project involves using ray-tracing tools to model the propagation of wideband signals in indoor and outdoor environments. Uninformed applications of such methods are computationally very costly.

We have developed a novel technique for reducing the computational burden associated with computing the electromagnetic fields associated with each ray over a wide bandwidth as well as novel visibility algorithms. These techniques could benefit further from application on a parallel architecture.

 Read more details about current projects in electromagnetic wave scattering here.

2. Envelope simulation of large scale circuits and systems

Our main present and future project will be the use of HPC for envelope simulation of large scale circuits and systems. Within this we will look at incorporating model reduction and partitioning tasks among various processors. The use of parallel computing and HPC would greatly aid the development of the most efficient envelope simulator.  Envelope simulators are used in RF simulation where complex modulated signals have widely varying timescales.  We are developing wavelet and time-domain envelope simulators and this is on-going work. In addition, ongoing research on Oscillator modelling will be extended. High-performance computing also benefits the development of simulation techniques for determining the nonlinear dynamics of coupled oscillators regardless of their application - biomedical or RF systems.

 Modelling in Astrophysics (Research Group of Turlough Downes)

3. Propagation of stellar jets

We are studying how well-collimated, narrow stellar jets can generate poorly collimated "molecular outflows" and what the observational properties of outflows generated in this way would be.  This involves running CFD simulations which incorporate chemistry and radiative losses appropriate to the astrophysical regime in question.  Both 2D (cylindrically symmetric) and full 3D simulations are required.

4. Turbulence in molecular clouds

 Turbulence in molecular clouds is thought to be a crucial ingredient in theories of star formation.  Stars form from these molecular clouds and the behaviour of any turbulence will influence how and what type of stars can form.  The study of fluid flow in molecular clouds is complicated by the presence of magnetic fields and the necessity of treating the difference components (neutral gas, dust and charged gas) individually, but coupling them through collisional terms.  This gives rise to a set of equations which have proved to be very challenging to solve numerically.  Recently, in collaboration with UCD (Dr Stephen O'Sullivan), we have derived a new numerical method which allows efficient solution of these equations in a parallel environment.  We are now applying the resulting code to the behaviour of turbulence.  This is extremely computationally demanding and requires significant computational resources (this is linked to an ICHEC project run by Dr Stephen O'Sullivan).

5. Decay of magnetic fields around neutron stars (with Dr Sarah Tanner (PDRA)

 Magnetic fields around neutron stars are very strong.  They are, however, weaker than would be expected from current theories of neutron star formation.  It is thought that the Hall effect may play a role in diffusing away some of the magnetic field generated during neutron star formation.  Using the code described in molecular clouds project we are addressing this issue.  This is not as computationally demanding as turbulence in molecular clouds, but still requires a parallel environment to satisfactorily study the physics involved.

6. Multifluid stability (with Dr Mohsen Shadmehri (PDRA)

 We are studying the Kelvin-Helmholtz instability in the context of flows in molecular clouds where magnetic fields and multifluid effects are important.  Preliminary analytic work has been done on this project.  The next step, which is imminent, is to run simulations of this situation and check for agreement with the analytic results in the first instance.  Once this has been achieved the full nonlinear behaviour of the instability can be studied.

To learn more about this projects and to see details of simulations, click here.

7. Stellar jet launching (funded under SFI RFP, hiring not complete yet).

 We will study how jets can be launched from the accretion disks surrounding forming stars.  This is an ambitious project requiring parallel computational resources at all stages.  It will involve full 3D simulations of the accretion disk, followed by simulations of the launching process and then, possibly, full-scale simulations of the entire system.

Theory and computation of Atomic, Molecular and Optical Processes (TCAMOP)  (Research Group of Lampros Nikolopoulos)

The development of theoretical and computational frameworks for the ab-initio handling of interaction of atomic and molecular systems in a strong radiation field is the main focus of our research. For this, an accurate solution of the time-dependent Schrodinger equation (TDSE) is absolutely necessary.

The solution of the TDSE is a challenging problem even for the simplest case: a hydrogen atom in a strong laser field. The study of the light-matter interactions with strong and ultra-short laser fields, beyond perturbation methods, was initiated about 25 years ago. At present, the advent of new light sources (NLS) with highly energetic photons, opens up new scientific territory. Free-Electron Laser (FEL) sources, delivering tunable and intense radiation in the VUV and XUV regimes, produce light capable of interacting directly with the inner-shell electrons of matter. Also, the development of ultra-short duration light pulses (durations on the sub-femtosecond time-scale) allows the exploration and control of the dynamics of physical processes occurring in the interiors of atoms and molecules.

Energy and Environmental Flow Modelling (Research group of Dr Yann Delaure, MME)

 The Energy and Environmental Flow Modelling Group was established in 2004 to foster research in Computational Fluid Dynamics modelling of complex heat and fluid flows found in energy systems and in particular Solar Thermal Collectors and Wave Energy Devices as well as a range of Environmental Flows. The group has developed a versatile and efficient two dimensional flow solver and has implemented a range of two and three dimensional models based on the OpenFOAM^® C++ libraries. It has extensive expertise in a range of commercial softwares and access to high performance computing resources. The group has been carrying a number of research project in both Energy and Environmental areas:

1.     Energy Flow

a.     Heat Transfer in Two-Fluid Flow

b.     Ocean Wave Interaction with Marine Structures 

c.     CFD Optimisation of Wind Turbine

2.     Environmental Flow

a.     The Effect of Fluid Flow on Bio-fouling

b.     Hydrodynamic and Ecological Modelling of Wastewater Discharge in Coastal Waters