Research
Below you can find some of the most interesting projects I carried out during my Master degree.
- Implementation of fireworks Python module, Plummer sphere evolution, code optimization Goal of this project is the development of a N-body simulation Python module. Using fireworks, we studied the evolution of different realization of star clusters drawn from a Plummer sphere in virial equilibrium. The cluster was evolved in orbit in a point mass potential. We analyzed the leading and trailing tidal arms and the 'tidal radius criterion' effectiveness in selecting the stars belonging to the cluster. Alongside with the simulation, we explored ways to optimize the N-body simulation code using GPUs and multi-threading and multi-processing on CPUs.
- Parallel N-body simulation using CUDA Goal of this project is the development of a parallel simulation of a 6-body gravitational problem in 3D space. The task involves computing the trajectories of six interacting particles under the influence of Newtonian gravity and visualizing their motion over time. The simulation is implemented using the CUDA framework.
- Hierarchical Mergers of Binary Black Holes Goal of this project is to understand the differences between hierarchical binary black hole mergers in NSCs, GCc and YSc, by looking at a set of simulated BBHs. Our analysis was carried out with classification ML algorithms, such as Random Forest and XGBoost. We proceed to analyze the importance of features to understand the properties of systems of BBHs.
- Dormant Black Holes in Binaries from Gaia DR3 We explored the evolution of binary systems containing a black hole and a main sequence star using Binary System Evolution (BSE) simulations and machine learning techniques. Using a rapid binary population-synthesis code, SEVN, we simulated the evolution of these systems and compared the results to the observed properties of known binaries. Machine learning techniques were employed to identify the key factors influencing the evolution of these systems.
- Streaming processing of cosmic rays using Drift Tubes detectors Goal of this project is to reproduce a real-time processing of real data collected in a particle physics detector and publish the results in a dashboard for live monitoring.
- Bayesian Blocks algorithm implementation in R Using the R program language, a Bayesian Block algorithm implementation has been developed and tested on light curve of Supernovae explosion.