Speaker
Description
Automated tuning is an area of active development at Rare Isotope Beam (RIB) facilities worldwide. Machine development tests at TRIUMF's ISAC (Isotope Separator and ACcelerator) facility have shown that automation can be achieved using combinations of simulation and machine learning models, categorizing tuning elements into those modeled by physics simulations and those addressing unknown deviations from that model.
This project aims to optimize ion beam extraction from the ECRIS "Supernanogan" source at TRIUMF-ISAC's Off-Line Ion Sources (OLIS) facility. We develop a new comprehensive simulation of plasma generation and the ion extraction system, alongside an optimization algorithm to maximize beam quality. A planned upgrade to a modern triode optics system necessitates this new model. By establishing a strong foundation for understanding beam formation, the model becomes the basis for a Bayesian optimization (BO) control algorithm: it provides good starting points and defines constraints on parameter-space exploration to ensure safety, whether expert-informed, simulation-informed, or model-learned.
Optimization will first address beam current transmission through a new collimator, translating into maximized beam brightness. A comprehensive variation of source parameters, including microwave heating power and gas pressure, will then follow.
| Classification | MC6: Applications and diagnostics |
|---|---|
| I have read and accept the Privacy Policy Statement | Yes |