Lieu : CEA-Saclay · Contrat : Stage · Rémunération : A négocier
The aim of the internship is to explore the possibility to use Artificial Intelligence (i.e. Machine Learning and Deep Learning Techniques) in order to predict the time evolution of the region of bounded motion (stability) in hadronic circular accelerators in presence of magnets imperfections. Usually numerical simulations are performed to compute the stability region, called Dynamic Aperture (DA), from particles motion in the accelerator model. These computations are time consuming, limiting the beam lifetime that can be simulated, in particular for large hadronic circular accelerators like LHC at CERN. Up to now, in order to overcome this limitation, only analytical models have been tried . The alternative way represented by the machine learning will be investigated in the proposed work.
The student will test the possibility to use Echo State Networks, employed successfully in the replication of chaotic attractors , or other Recurrent Neural Networks to predict the long-term stability region in the cases of future upgrades of LHC.
 A. Bazzani et al. “Advances on the modeling of the time evolution of dynamic aperture of hadron circular accelerators”, Phys. Rev. Acc. and Beams 22, 104003 (2019).
 J. Pathak et al. “Using Machine learning to replicate chaotic attractors, and calculate lyapunov exponent from data”, chaos 27, 121102 (2017).
Graduate course Master or Engineer with good knowledge in Machine Learning techniques, analytical calculations and affinity for theoretical studies.