Slides of the C++ course which I gave at Cineca supercomputing center, Italy. pdf (>100 slides)
Presentation at SuperComputing05, Seattle, US 2005. pdf (>10MB)
Presentation for the meeting "Particle methods and mesoscale simulation of liquids" at Daresbury Laboratory 2005. pdf
MyDPD
MyDPD is a mesodynamics code in three spatial dimensions, where DPD stands for dissipative particle dynamics. This code is simple but functional. It contains two integrators for the DPD stochastic equations, a simple DPD velocity Verlet and the stochastic Trotter integrators.
Foundations of dissipative particle dynamics (DPD)
- I have been working on the theoretical foundation of DPD in terms of the relation with the fluctuating hydrodynamics (FH) equations. We derived, using kinetic theory and the formalism of the Fokker-Planck equation, the mesoscopic anologous of DPD based on a Lagrangian finite volume discretization of FH over the Voronoi tessellation. We developped a code which create dynamic periodic Voronoi tessellations in 3-dim to be used with the method.
For a presentation look here:Voronoi for complex fluids [pdf]; Here a movie of a dynamic periodic Voronoi tessellation (>4MB)
Lattice gas and lattice Boltzmann methods
Von Karmen vortex movie with LBE(>3MB)
Growth and growth distributions
- We investigated possible probabilistic explanations of the negative relationship between firm size and the variability of growth rates, in order to lay some groundwork for a more careful modelling of firm dynamics. Our results are based on a detailed database on the pharmaceutical industry.
We have also proposed a variety of models of random walk, discrete in space and time, suitable for simulating stable random variables of arbitrary index (0 < 2). We show that by properly scaled transition to vanishing space and time steps our random walk models converge to the corresponding continuous Markovian stochastic processes, that we refer to as L´evy-Feller diffusion processes. (see publications).