Cluster analysis for granular mechanics simulations using Machine Learning Algorithms

RIM, DANIELA N.; MILLÁN, EMMANUEL N.; PLANES, BELÉN; BRINGA, EDUARDO; LUIS G. MOYANO

Mendoza

Congreso; Congreso Internacional de Ciencias de la Computación y Sistemas de Información; 2018

UNIVERSIDAD CHAMPAGNAT

Molecular Dynamics (MD) simulations on grain col- lisions allow to incorporate complex properties of dust interac- tions. We performed simulations of collisions of porous grains, each with many particles, using the MD software LAMMPS. The simulations consisted of a projectile grain striking a larger immobile target grain, with different impact velocities. The dis- advantage of this method is the large computational cost due to a large number of particles being modeled. Machine Learning (ML) has the power to manipulate large data and build pre- dictive models which could reduce MD simulation times. Using ML algorithms (Support Vector Machine and Random Forest) we are able to predict the outcome of MD simulations regard- ing fragment formation, after a number of steps smaller than in usual MD simulations. We achieved a time reduction of at least 46%, for 90% accuracy. These results show that SVM and RF can be powerful yet simple tools to reduce computational cost in collision fragmentation simulations.