Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques
<jats:p>We present a novel machine learning-based approach to generate fast-executing virtual radiofrequency quadrupole (RFQ) particle accelerators using surrogate modelling. These could potentially be used as on-line feedback tools during beam commissioning and operation, and to optimize the...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Frontiers Media SA
2022
|
| Online Access: | https://hdl.handle.net/1721.1/142240 |
| _version_ | 1811069113625214976 |
|---|---|
| author | Koser, Daniel Waites, Loyd Winklehner, Daniel Frey, Matthias Adelmann, Andreas Conrad, Janet |
| author2 | Massachusetts Institute of Technology. Laboratory for Nuclear Science |
| author_facet | Massachusetts Institute of Technology. Laboratory for Nuclear Science Koser, Daniel Waites, Loyd Winklehner, Daniel Frey, Matthias Adelmann, Andreas Conrad, Janet |
| author_sort | Koser, Daniel |
| collection | MIT |
| description | <jats:p>We present a novel machine learning-based approach to generate fast-executing virtual radiofrequency quadrupole (RFQ) particle accelerators using surrogate modelling. These could potentially be used as on-line feedback tools during beam commissioning and operation, and to optimize the RFQ beam dynamics design prior to construction. Since surrogate models execute orders of magnitude faster than corresponding physics beam dynamics simulations using standard tools like PARMTEQM and RFQGen, the computational complexity of the multi-objective optimization problem reduces significantly. Ultimately, this presents a computationally inexpensive and time efficient method to perform sensitivity studies and an optimization of the crucial RFQ beam output parameters like transmission and emittances. Two different methods of surrogate model creation (polynomial chaos expansion and neural networks) are discussed and the achieved model accuracy is evaluated for different study cases with gradually increasing complexity, ranging from a simple FODO cell example to the full RFQ optimization. We find that variations of the beam input Twiss parameters can be reproduced well. The prediction of the beam with respect to hardware changes, e.g., the electrode modulation, are challenging on the other hand. We discuss possible reasons for that and elucidate nevertheless existing benefits of the applied method to RFQ beam dynamics design.</jats:p> |
| first_indexed | 2024-09-23T08:05:55Z |
| format | Article |
| id | mit-1721.1/142240 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T08:05:55Z |
| publishDate | 2022 |
| publisher | Frontiers Media SA |
| record_format | dspace |
| spelling | mit-1721.1/1422402023-01-30T21:33:50Z Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques Koser, Daniel Waites, Loyd Winklehner, Daniel Frey, Matthias Adelmann, Andreas Conrad, Janet Massachusetts Institute of Technology. Laboratory for Nuclear Science <jats:p>We present a novel machine learning-based approach to generate fast-executing virtual radiofrequency quadrupole (RFQ) particle accelerators using surrogate modelling. These could potentially be used as on-line feedback tools during beam commissioning and operation, and to optimize the RFQ beam dynamics design prior to construction. Since surrogate models execute orders of magnitude faster than corresponding physics beam dynamics simulations using standard tools like PARMTEQM and RFQGen, the computational complexity of the multi-objective optimization problem reduces significantly. Ultimately, this presents a computationally inexpensive and time efficient method to perform sensitivity studies and an optimization of the crucial RFQ beam output parameters like transmission and emittances. Two different methods of surrogate model creation (polynomial chaos expansion and neural networks) are discussed and the achieved model accuracy is evaluated for different study cases with gradually increasing complexity, ranging from a simple FODO cell example to the full RFQ optimization. We find that variations of the beam input Twiss parameters can be reproduced well. The prediction of the beam with respect to hardware changes, e.g., the electrode modulation, are challenging on the other hand. We discuss possible reasons for that and elucidate nevertheless existing benefits of the applied method to RFQ beam dynamics design.</jats:p> 2022-05-03T12:33:16Z 2022-05-03T12:33:16Z 2022-04-25 Article http://purl.org/eprint/type/JournalArticle 2296-424X https://hdl.handle.net/1721.1/142240 Koser, Daniel, Waites, Loyd, Winklehner, Daniel, Frey, Matthias, Adelmann, Andreas et al. 2022. "Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques." 10. 10.3389/fphy.2022.875889 Creative Commons Attribution 4.0 International License https://creativecommons.org/licenses/by/4.0 application/pdf Frontiers Media SA Frontiers |
| spellingShingle | Koser, Daniel Waites, Loyd Winklehner, Daniel Frey, Matthias Adelmann, Andreas Conrad, Janet Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques |
| title | Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques |
| title_full | Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques |
| title_fullStr | Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques |
| title_full_unstemmed | Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques |
| title_short | Input Beam Matching and Beam Dynamics Design Optimizations of the IsoDAR RFQ Using Statistical and Machine Learning Techniques |
| title_sort | input beam matching and beam dynamics design optimizations of the isodar rfq using statistical and machine learning techniques |
| url | https://hdl.handle.net/1721.1/142240 |
| work_keys_str_mv | AT koserdaniel inputbeammatchingandbeamdynamicsdesignoptimizationsoftheisodarrfqusingstatisticalandmachinelearningtechniques AT waitesloyd inputbeammatchingandbeamdynamicsdesignoptimizationsoftheisodarrfqusingstatisticalandmachinelearningtechniques AT winklehnerdaniel inputbeammatchingandbeamdynamicsdesignoptimizationsoftheisodarrfqusingstatisticalandmachinelearningtechniques AT freymatthias inputbeammatchingandbeamdynamicsdesignoptimizationsoftheisodarrfqusingstatisticalandmachinelearningtechniques AT adelmannandreas inputbeammatchingandbeamdynamicsdesignoptimizationsoftheisodarrfqusingstatisticalandmachinelearningtechniques AT conradjanet inputbeammatchingandbeamdynamicsdesignoptimizationsoftheisodarrfqusingstatisticalandmachinelearningtechniques |