PRACE-ICEI – Call #1 Awarded Projects

On this page you will find the projects that were awarded under PRACE-ICEI Calls For Proposals – Call #1.

Awarded Projects

Project Title: High pressure thermal conductivity of β-Ga2S3

Project Leader: Mr Samuel Gallego Parra, Technical University of Valencia, Spain

Resource Awarded
The following resources were awarded – hosted by ETH Zurich/CSCS, Switzerland:

  • 1 000 node hours
  • 69 600 node hours per quarter on Piz Daint
  • 3 TB of storage

Research Field: Materials Science

Abstract:
The goal of the project is the search of new thermoelectric materials (TMs) by considering high pressure (HP) methods. Indeed, application of HP has allowed not only the enhancement of thermoelectric properties, but also the possibility of obtaining new structural phases which exhibit these properties by means of pressure-induced phase transitions. A major interest within the HP community is the search for systems which evidence pressure-induced thermoelectric properties and are composed by lighter and eco-friendly constituents.

An interesting example is Ga2S3, a compound that is yet to be explored in depth. Ga2S3 undergoes a phase transition from the α’ (space group (S.G.) Cc) to the β phase (S.G. R3 ̅m) at around 16 GPa (X. Lai et al., J. Appl. Phys., 2014, 116, 193507). Concerning to Ga2S3, our group has carried out HP X-ray diffraction and HP Raman scattering measurements to revisit the previously published work, as we have done in other similar TMs (A.L. Pereira et al., J. Phys. Chem. C, 2018, 122, 8853-8867). The ICEI award will help in obtaining further information regarding the thermal conductivity performance as a function of pressure, providing the starting point to explore β-Ga2S3 as an emerging TM.

Project Title: ACOUPROP: Hydro-acoustic characterization of a marine propeller

Project Leader: Dr. Antonio Posa, CNR-INM, Institute of Marine Engineering, Italy

Resource Awarded
The following resources were awarded – hosted by CINECA, Italy:

  • 25 000 node hours of interactive computing
  • 10 TB active storage
  • 200 archive storage

Research Field: Fluid dynamics, Computational engineering, Civil engineering, Maritime/hydraulic engineering, Geotechnics, Waste treatment

Abstract:
Noise generated by propellers is a major issue in naval transportation. For instance, it has a substantial environmental impact, influencing especially the orientation strategies adopted by fishes and marine mammals. It also negatively affects the comfort of passengers and crew members on board of ships. Nonetheless, the scientific literature on hydro-acoustics of propellers is very limited, which is a direct consequence of the experimental and computational challenges associated to this subject. In addition, recent studies point that assumptions and approximations largely adopted in aero-acoustics, which is a better-known research topic, do not apply in hydro-acoustics, making its study even more demanding. In the present project we aim at analysing the large database generated by high-fidelity computations in the framework of our earlier PRACE projects #2016163889 and #2018184409 (15th and 17th Calls for Project Access) with the purpose of extracting new insight on the acoustic signature of marine propellers, both in isolated conditions and in presence of upstream and downstream devices. This analysis will enable us to identify the main sources of noise and to provide guidance about the fidelity of the different approaches and approximations utilized in the field to assess the acoustic signature in marine propulsion.

Project Title: HydroTurb: Wake analysis of a hydrokinetic turbine

Project Leader: Dr. Riccardo Broglia, CNR-INM, Institute of Marine Engineering, Italy

Resource Awarded
The following resources were awarded – hosted by GENCI/CEA, France:

  • 1 000 node hours of interactive computing
  • 10 000 TBxday active storage
  • 150 archive storage

Research Field: Fluid dynamics, Computational engineering, Energy systems

Abstract:
The kinetic energy of tides, rivers and marine currents can provide an important contribution in replacing fossil fuels, being abundant, renewable and clean. The environmental impact of its exploitation is significantly smaller than conventional technologies. Therefore, although its utilization is currently very limited, one can envision a significant expansion during the next years. Axial-flow hydrokinetic turbines, whose development comes mainly from the wind energy industry, represent one of the major technologies for harvesting energy from tides, rivers and seas. As typical in wind energy, one major issue consists in characterizing the wake signature of such turbines, with the purpose of defining the spacing between devices, in order for the downstream ones to operate with enough inflow kinetic energy for efficient performance. In this study the database generated in the framework of the on-going PRACE project #2019204935 (19th Call for Project Access) will be utilized to analyse the wake of an axial-flow hydrokinetic turbine for three different values of rotational speed. This analysis will be utilized for visualization of the instability phenomena of the wake, which promote momentum replenishment downstream, and for assessing the influence of the rotational speed on turbulence statistics and the process of momentum recovery within the wake.

Project Title: Thermodynamic and Dynamical Stability of Hf and Zr Based Ruddlesden-Popper Structures

Project Leader: Dr. Estelina Lora Da Silva, University of Porto, Portugal

Resource Awarded
The following resources were awarded – hosted by CINECA, Italy:

  • 75 456 node hours of interactive computing
  • 50 TB archive storage

Research Field: Materials Science/Physics

Abstract:
Ruddlesden-Popper (RP) structures (An+1BnO3n+1 stoichometry) are layered-type perovskite materials, and can be seen as stackings of ABO3 perovskite blocks, with the extra AO rock-salt sheet intercalated between every nth perovskite unit. The perturbation of the AO sheet will inducenonpolar structural distortions to the perovskite system, e.g. BO6 octahedral rotations and tiltings, allowing however breaking of the inversion symmetry and giving rise to spontaneous polarization. These materials present a wide variety of applications, but here the main interest focuses on RP structures as emergent energy-efficient electronic devices, i.e. ferroelectric random access memory and high-efficient photovoltaics [Energy Environ. Sci. 8, 838 (2015). Nat. Commun. 4, 1990 (2013)]. By considering the reference high-symmetry I4/mmm and polar low-symmetry Cmc21 systems, which is the ferroelectric ground-state for the prototype structure of both Ca3Ti2O7 and Ca3Mn2O7 [J. Solid State Chem. 195, 11 (2012)], we intend to employ evolutionary algorithms together with the density functional framework to analyse the A3B2O7 systems. We intend to perform a complete screening of the configurational space for systems with different combinations of chemical species at the A- and B-sites, and analyse the relation between tolerance factor with respect to the thermodynamic and dynamical stability of the different compositional structural phases.

Project Title: BigHeart: big data analysis and simulation of human heart function

Project Leader: Prof. Alfonso Bueno-Orovio, University of Oxford, United Kingdom

Resource Awarded
The following resources were awarded – hosted by ETH Zurich/CSCS, Switzerland:

  • 1 000 node hours
  • 35 050 node hours per quarter on Piz Daint
  • 100 TB of storage

Research Field: Computational Medicine

Abstract:
Cardiovascular disease remains the major cause of death worldwide, led by myocardial infarction and inherited cardiomyopathies. The aging of our population is further aggravating this societal burden, with a sustained increase of more than a million of deaths per year. These figures highlight the need of developing new Computational Medicine approaches, to complement clinical research in order to improve the understanding and treatment of heart disease.

This application aims at generating the first large-scale dataset of patient-specific anatomical models and validated simulations of human electromechanical cardiac function, in both health and disease. The later will focus on myocardial infarction and hypertrophic cardiomyopathy, as leading causes of mortality among the elderly and the young, respectively. ICEI resources will enable these ambitious goals, by providing an integrated e-research infrastructure to optimise our present workflow, currently highly fragmented between different types of desegregated computing services.

The results obtained during this project will provide novel insights on mechanical dysfunction in human heart disease. Their distinct large-scale nature will also establish the basis for their future exploitation.

These are expected to push the boundaries of in silico human clinical trials for drug and electrical therapy, and the realisation of the digital twin in clinical decision-making.

Project Title: Calibration of the paramters of a network model for studying the transmision dynamics of the SARS-Cov-2

Project Leader: Dr. Rafael J Villanueva, Universitat Politècnica de Valéncia, Spain

Resource Awarded
The following resources were awarded – hosted by GCS at JSC, Germany:

  • 5 860 node hours of interactive computing

Research Field: Modeling, Mathematical epidemiology

Abstract:
The main goal of this proposal is to build a network model to study the transmission dynamics of the SARS-Cov-2. This model is completely computational, building big (more than 1 million nodes) random networks where we simulate how the virus spreads.

A network consists of nodes links between pairs of individuals, representing contacts (effective or not) for disease transmission. On a network, using computer programs, it is possible to simulate the evolution of the transmission of an infectious disease over the time. In network models we can follow any individual and implement health policies by selecting them according to their clinical record. In our group, we have studied random networks in epidemiology [1, 2].

References:
[1] Acedo, L.; Moraño, J.A.; Villanueva, R.J.; Villanueva-Oller, J.; Díez-Domingo, J. Using random networks to study the dynamics of respiratory syncytial virus (RSV) in the Spanish region of Valencia. Math. Comp. Mod. 2011, 54, 1650–1654. https://doi.org/10.1016/j.mcm.2010.11.068
[2] Díez-Domingo, J.; Sánchez-Alonso, V.; Villanueva, R.-J.; Acedo, L.; Moraño, J.-A.; Villanueva- Oller, J. Random Network Models to Predict the Long-Term Impact of HPV Vaccination on Genital Warts. Viruses 2017, 9, 300. https://doi.org/10.3390/v9100300