Beam Tuning Automation Activities at TRIUMF

Autor: Kiy, Spencer, Adegun, Joseph, Ames, Friedhelm, Andres, Achim, Baartman, Richard, Bagri, Harpriya, Ezawa, Keiko, Fedorko, Wojtek, Jung, Paul, Kester, Oliver, Lucow, Kevin, Nasser, Jamiel, Planche, Thomas, Rädel, Stephanie, Schultz, Brad, Shelbaya, Olivier, Stringer, Blair, Thomson, Daniel, Wang, David, Wu, Kristin
Jazyk: angličtina
Rok vydání: 2022
Předmět:
DOI: 10.18429/jacow-hiat2022-tu2c4
Popis: The particle accelerator complex at TRIUMF provides beams for secondary particle production including rare isotopes. The post acceleration of rare isotope ions demands frequent changes of beam properties like energy and changes of the ion species in terms of isotope and charge state. To facilitate these changes to beam properties and species, a High Level Applications (HLA) framework has been developed that provides the essential elements necessary for app development: access to sophisticated envelope simulations and any necessary beamline data, integration with the control system, version control, deployment and issue tracking, and training materials. With this framework, one can automate collection of beam data and subsequently pull that data into a model which then outputs the necessary adjustments to beam optics. Tuning based on this method is model coupled accelerator tuning (MCAT) and includes pursuits like the training of machine learning (ML) agents to optimize corrections benders. A summary of the framework will be provided followed by a description of the different applications of the MCAT method - both those currently being pursued, and those envisioned for the future.
Proceedings of the 15th International Conference on Heavy Ion Accelerator Technology, HIAT2022, Darmstadt, Germany
Databáze: OpenAIRE