Speaker
Description
The application of Artificial Intelligence (AI) and Machine Learning (ML) in particle accelerator systems has become an effective strategy for handling complex operations and enhancing performance. At INFN-Legnaro National Laboratories (INFN-LNL), both offline and online AI/ML-driven approaches have been developed to improve beam dynamics, reduce setup times, and increase overall accelerator efficiency.
Offline efforts focus on surrogate modeling of complex facilities such as ANTHEM BNCT and PIAVE-ALPI, as well as on virtual diagnostics implemented through supervised neural networks. By combining these tools with AI/ML optimization algorithms, new design and commissioning strategies are being studied to further enhance beam quality and operational performance.
In parallel, online real-time optimization strategies using Bayesian algorithms and Particle Swarm Optimization (PSO) have delivered promising results. Notably, at the PIAVE-ALPI superconducting accelerator, the use of BO improved beam transmission to 85%, a remarkable increase from the typical 35% operational average. Together, these advances demonstrate the growing impact and future potential of AI/ML technologies in accelerator science and operations.
