projects
POCI-01-0145-FEDER-028118
POCI-01-0145-FEDER-028118
ESC4SHI - Simulação e Computação eficiente para a Saúde, o Mar e a Indústria
ESC4SHI - Simulação e Computação eficiente para a Saúde, o Mar e a Indústria
17/12/2018-16/12/2021. Funding 240.000,00€
Main researcher: S. Clain
FCT-ANR/MAT-NAN/0122/2012
FCT-ANR/MAT-NAN/0122/2012
New numerical schemes for extreme phenomena simulation
New numerical schemes for extreme phenomena simulation
Main researcher: S. Clain
01/03/2013--31/03/2016.
Funding: 72.460,00€.
EXPL/MAT-NAN/0606/2013
EXPL/MAT-NAN/0606/2013
VAPS- Abordagem variacional para Cirurgia Plástica.
VAPS- Abordagem variacional para Cirurgia Plástica.
01/01/2014--30/06/2015. Funding: 28.750,00 e
Main researcher: Graça Maria Reizinho Carita.
Simulation coordinator: S. Clain
PTDC/MAT/121185/2010
PTDC/MAT/121185/2010
Numerical Simulation of three-dimensional polymer flow and application to molding processes
Numerical Simulation of three-dimensional polymer flow and application to molding processes
Main researcher: S. Clain
02/04/2012--01/08/2015.
Funding: 70.000,00 €
Research lines
Research lines
I manage several topics in scientific computing and machine learning
I manage several topics in scientific computing and machine learning
- Very high order numerical scheme: Development of numerical scheme of order higher than three, boundary condition on curved interface, MOOD method for hyperbolic system.
- Computational efficiency and HPC: New formulations for very high order finite difference method to take advantage of the parallelism, vectorization, cache memory and contiguous data. Simulation on complex domains with structured grids.
- Modelling for environmental problems: Very high order shallow water solver for tsunami simulation, coastal protection model.
- Meshless methods (SPH, ...): Consistent SPH scheme, very high order SPH scheme, MOOD for SPH
- Machine Learning and data sciences: Machine learning for numerical simulation, predictor-corrector, polynomial interpolation with autoencoder, utility clustering.