Antonis Papadakis

Associate Professor
Frederick University

Disciplines, scientific fields, research areas

  • Engineering Sciences & Technology
    • Computer and telecommunications engineering
      • Computational methods in engineering
      • Computer engineering
      • Information and intelligent systems engineering
    • Electrical, electronic & communication engineering
      • Computer hardware and architecture
    • Environmental engineering & biotechnology
      • Environmental engineering
    • Other Engineering Sciences and Technology
  • Environment & Energy
    • Circular economy
      • Sustainable industry and manufacturing systems
    • Climate change
      • Modelling and projections
    • Earth and related environmental sciences
      • Atmospheric sciences
    • Energy and the built environment
      • Energy technologies for buildings
    • Renewable energy resources and systems
      • Wind energy

Keywords

  • AI
  • CFD
  • Green Technologies
  • Graphical User Interface
  • Software developer
  • Computer Aided Engineering
  • Optimization
  • Engineering Systems
  • Energy and Buildings

Bio

Dr. Papadakis has graduated with a PhD from Cambridge University, on a full-time 4-year scholarship from the Department of Engineering, Trinity College (Trinity staff hold 34 Nobel Prizes and 5 Fields medals), UK in 2004. Thereafter he worked as a Postdoctoral Fellow in the CMS experiment at CERN for 3 years conducting Monte-Carlo simulations and was a member of the CMS group that discovered the Higgs particle in 2012, which resulted in a Nobel Prize winning award in 2013. Dr. Papadakis is the CEO and founder of KYAMOS, and an Associate Professor at Frederick University Cyprus, Department of Electrical Engineering. He was also the Vice-Chairman of the Department of Electrical Engineering and currently is the Coordinator of the MSc in Oil & Gas and Offshore Engineering program. He is an author in 58 peer-reviewed journal publications and 36 international conference publications, with more than 23,500 google scholar citations and an h-index of 33. He has authored 2 Book chapters in the field of plasma physics, and he has conducted extensive numerical and experimental electromagnetic studies for overhead and underground power lines, and open and closed substations in Cyprus. Dr. Papadakis has solid expertise in the field of numerical modelling. He is also a reviewer for 14 international peer-reviewed journals from ELSEVIER, IET, IEEE, IOP and AIP in the areas of plasma, electromagnetics, and high-performance computing. Three of his journal papers (two from the Institute of Physics, and one from Elsevier) have received honorary mention from the journals. He has been a steering committee member of DEMSEE International Conference since 2011 and acted as Co-Chairman of the DEMSEE2014 and Chairman of the DEMSEE2018 International Conference held in September 2018 in Cyprus. He has also participated as Cyprus delegate into two COST-actions, one related to lightning discharges and one on liquid-plasma interaction. In the past, Dr. Papadakis has coordinated 2 Cyprus RPF projects in collaboration with CERN, one as Project Coordinator (power electronic devices) and one as Scientific Coordinator. Recently, he Coordinated a Horizon 2020 Teaming Phase 1 project on behalf of Frederick University of €400,000. He also received funding and successfully coordinated the Cyprus participation in the EURATOM-ITER project through the Hellenic Association of Euratom from 2005-2008, conducting plasma-fluid numerical simulations. Finally, he has participated in an incubator programme for Live-Fingerprinting recognition hardware in 2010 worth of €200,000. He was appointed in the Technical Chamber of Cyprus for registering members to the electrical engineering occupation for 10 years in Cyprus. He has also been appointed and acted as an expert through ETEK regarding the installation of a 132-kV overhead power line from the Electricity Authority of Cyprus above the field of his client, investigating the possibility of hazardous flashover while farming, which resulted in out of the court settlement on behalf of his client of €80,000 worth. Recently, Dr. Papadakis has secured several funding opportunities. Specifically, (a) €35,000 from the European Space Agency (ESA) to conduct high performance computing simulations utilizing the CUDA aware MPI technology in the field of astrophysics for the discovery of new stars and galaxies, (b) funding through the Erasmus+ for the exchange of 5 PhD students and 2 member staff from Pakistan (NUST and COMSATS Universities), (c) he has successfully graduated from the 9th month IDEA business training program and has secured a total of 7 grants (€62,000, €41,060, €96,692, €35,700, €50,120, €736,105, €51,520), acting as Principal Investigator, since the registration of the KYAMOS company on the 13th July 2018. Finally, Dr. Papadakis is also participating in an awarded 2023 €388,000 DUAL USE/0922/0055 project of the Research and Innovation Foundation, in his capacity as Frederick Researcher.

Degrees

PhD in Electrical Engineering (1999–2003) - PhD in Modeling of Gas Discharges using Finite Elements: Incorporation of Navier-Stokes equations
Department of Electrical Engineering, Cambridge University, Trinity College, Cambridge (United Kingdom)


BENG in Electrical Engineering (1996–1999), Grade: First Class Honors
Department of Electrical Engineering, University of Warwick, Coventry (United Kingdom)

Projects

KYAMOS Software–Cloud GPU InfiniBand tool for modeling COVID19, Authors: A. P. Papadakis, A. Ioannou, A. Georgiou, and W. Almady

01/01/2021 - 01/10/2021
In this paper, we present an introduction to the Coronavirus transmission behaviour between humans. We then present a literature review of the research conducted with regards to transmission and propagation of the Coronavirus using computational fluid dynamics. We give special emphasis on the methods employed and most importantly their speed and accuracy. Thereafter, we present the theory behind Coronavirus transmission and explain how to model Coronavirus propagation. A mathematical model capable of capturing the corona particle propagation, and specifically the infective aerosols attached to the moving air particles is presented together with its mathematical formulation. We explain thoroughly the Lattice-Boltzmann Multiple-Relaxation- Time, Smagorinsky Large-Eddy-Simulation model used to capture the sub-grid turbulent dynamics, and finally the model, implemented in CUDA, is discussed and the results of the Coronavirus propagation inside a room are presented. We investigate different realistic scenarios such as having two persons in a room breathing and talking and we produce different aerosol concentration plots and air particle speed velocities.

KYAMOS Software Finite Volume TVD scheme for advection fluid simulations, Authors: A. Papadakis, A. Ioannou and W. Almady

02/12/2019 - 02/12/2020
The Computer Aided Engineering industry utilizes computer simulations to analyze and predict mainly fluid flow phenomena. These simulations can be used both for research and commercial purposes, to comprehend physical phenomena, that will aid in the design of engineering systems. There are various
technologies out there, that can be used to perform computing simulations, which are well-established. In this paper, firstly, the various new
technologies that are expected to produce a disrupt in the market in the near future are discussed which are artificial intelligence, quantum computing and tensor processing units. Thereafter, the various methods to approximate
partial differential equations are analyzed, and focus is given in the solution of convection-dominated problems, which is a major part of fluid simulation phenomena. We therefore discuss the advantages and disadvantages of the main
traditional methods such as the finite difference, finite element and finite volume methods, as well as more innovative methods such as the Lattice
Boltzmann method in mesoscopic range. In this paper, a very innovative Finite Volume-Total Variation Diminishing scheme is proposed, that has the capability to ideally predict the fluid flow of both diffused and shock wave phenomena, free from artificial numerical diffusion, spurious oscillations and guaranteeing flux conservation. The results demonstrate that the proposed
method captures the waves in a nearly ideal way and guarantees accurate simulations, even when long time simulations are performed, excelling in
accuracy. Since it is developed in a CUDA aware MPI environment, it is highly computationally efficient and it is expected to disrupt the market and establish KYAMOS software, as a competitive alternative in the Computer Aided Engineering industry.

KYAMOS Software – CUDA aware MPI Solver for Poisson equation, Authors: Antonis P. Papadakis, Aimilios Ioannou and Wasif Almady

N/A - 01/12/2020
In this paper, we first conduct a brief introduction to computational fluid dynamics simulations, and discuss the different types of differential equations used in the macroscopic world. Then we discuss the theory behind adaptive mesh techniques, its advantages and disadvantages and present an algorithm that
produces nearly ideal mesh element qualities. Then we describe the finite element Galerkin formulation of the Poisson equation using an iterative Conjugate Gradient method and its parallelization into CUDA aware MPI. Finally, we conduct validation results for the Poisson equation in uniform geometries and test the scaling of our computer software to multiple threads and GPUs.

KYAMOS Multiphysics Software - Lattice Boltzmann Solver, Author: A. Papadakis

02/12/2019 - 01/11/2020
In this paper, we describe KYAMOS software attributes, as well as the development of a new innovative Lattice Boltzmann equation solver developed
within an InfiniBand distributed GPU computing cluster. Lattice Boltzmann solvers are well known for treating boundary conditions with ease and are
highly suitable for parallel processing. Hence, an InfiniBand GPU cluster is customary developed in operating system Linux/CENTOS. Its purpose is through cluster computing, to provide considerable computational power to every client to solve large problems with the desired speed and accuracy, having deployed a very friendly Graphical User Interface environment. Most importantly, KYAMOS software is based on in house developed algorithms and ready-made, well tested modules. Benchmark results for the time dependent, lid driven cavity flow test case is presented and discussed. It is found that the GPU based LB solver is approximately 25 times faster than an identical serial solver developed using competitor’s software.

Trivia about me

Dr. Papadakis is the CEO and founder of KYAMOS LTD, and an Associate Professor of the Department of Electrical Engineering at Frederick University Cyprus. He was also the Vice-Chairman of the Department of Electrical Engineering and is the Coordinator of the MSc in Oil & Gas and Offshore Engineering program. He is an author in 58 peer-reviewed journal publications and 36 international conference publications, with more than 23,500 google scholar citations and an h-index of 33. He has authored 2 Book chapters in the field of plasma physics, and he has conducted extensive numerical and experimental electromagnetic studies for overhead and underground power lines, and open and closed substations in Cyprus. He is currently working on a CODEVELOP project, namely “REAl LIve SimulATION of GREEN technologies in the CLOUD”, with a funding amount of €736.105 that will be completed on 28th February 2025.