Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Tuning of the ESS Drift Tube Linac using Machine Learning

Lundquist, Johan LU (2022) FYSM60 20212
Department of Physics
European Spallation Source ESS AB
Abstract
The European Spallation Source, currently under construction in Lund, Sweden, will be the world's brightest neutron source. It is driven by a linear accelerator designed to accelerate a beam of protons with 62.5 mA, 2.86 ms long pulses, working at 14 Hz. The final section of its normal-conducting front-end consists of a 39 m long drift tube linac divided into five tanks, designed to accelerate the proton beam from 3.6~MeV to 90 MeV. The high beam current and power impose challenges to the design and tuning of the machine. In order to keep the beam quality throughout the accelerator and beam losses at a minimum the radio frequency amplitude and phase within the accelerating components have to be set within 1% and 1 deg of the design values.... (More)
The European Spallation Source, currently under construction in Lund, Sweden, will be the world's brightest neutron source. It is driven by a linear accelerator designed to accelerate a beam of protons with 62.5 mA, 2.86 ms long pulses, working at 14 Hz. The final section of its normal-conducting front-end consists of a 39 m long drift tube linac divided into five tanks, designed to accelerate the proton beam from 3.6~MeV to 90 MeV. The high beam current and power impose challenges to the design and tuning of the machine. In order to keep the beam quality throughout the accelerator and beam losses at a minimum the radio frequency amplitude and phase within the accelerating components have to be set within 1% and 1 deg of the design values.

One of the usual methods used to define the radio frequency set-point is called signature matching, which can be a challenging process, and new techniques to meet the growing complexity of accelerator facilities are highly desirable. Machine learning is a rapidly growing field which has found applications in a wide range of scenarios, accelerators being no exception, but the tuning of RF fields using machine learning has yet to be tried. This project explores the possibility of applying machine learning in this area of accelerator physics, comparing this novel technique with the established signature matching and introducing a new possibility for faster tuning using a different data structure.

For this purpose, simulations of the first tank in the drift tube linac section of the ESS linac were used to produce large amounts of tuning data at varying setpoints for the machine. Data like this was then used with the signature matching and machine learning techniques to fit the necessary functions for the traditional technique and train artificial neural networks for the novel technique. Random machine errors were later introduced to test each method's generalized performance. This data was also restructured to allow for machine tuning in a single shot, while usually a parameter scan is necessary, and machine learning techniques were tried for tuning on this data.

Machine learning has been found to perform well in comparison with the established method, with some select advantages inherent to machine learning. This faster RF tuning technique only possible with machine learning, is found to perform well, although not quite within the given 1% and 1 deg limitations. As the rest of the results of this project are all on simulated data for the ESS, a comparison with results of these techniques on real data collected from the Spallation Neutron Source in the USA was performed. Future improvements could include workarounds for faulty network inputs and further tuning of the networks used. (Less)
Popular Abstract (Swedish)
European Spallation Source (ESS) är en världsledande anläggning för forskning med hjälp av neutroner, under konstruktion i Lund. För att producera neutroner för sådan forskning krävs kraftigt accelererade protoner. Sådan acceleration måste utföras mycket exakt och säkert, och för detta krävs exakta kalibreringsmetoder. I detta projekt har en undersökning utförts huruvida maskininlärning kan vara till hjälp under dessa kalibreringar.

Processen via vilken ESS får ut sina neutroner kallas för spallation. För att denna kärnreaktion ska ske krävs det att protoner kolliderar med ett tungt ämne i hastigheter nära ljusets hastighet. På ESS utnyttjas en disk av volfram som kollisionsmål, och de extremt snabba protonerna förses via en linjär... (More)
European Spallation Source (ESS) är en världsledande anläggning för forskning med hjälp av neutroner, under konstruktion i Lund. För att producera neutroner för sådan forskning krävs kraftigt accelererade protoner. Sådan acceleration måste utföras mycket exakt och säkert, och för detta krävs exakta kalibreringsmetoder. I detta projekt har en undersökning utförts huruvida maskininlärning kan vara till hjälp under dessa kalibreringar.

Processen via vilken ESS får ut sina neutroner kallas för spallation. För att denna kärnreaktion ska ske krävs det att protoner kolliderar med ett tungt ämne i hastigheter nära ljusets hastighet. På ESS utnyttjas en disk av volfram som kollisionsmål, och de extremt snabba protonerna förses via en linjär accelerator. Denna över 500 meter långa accelerator ligger under jorden och utnyttjar en mångfald av tekniker och komponenter för att accelerera protonerna. Figuren nedan visar en enkel modell av ESS acceleratorn, med var särskild avdelning markerad med ett förkortat namn för komponenten. Alla dessa komponenter bidrar till protonernas slutgiltiga hastighet: över 80 % av ljushastigheten.

När protonerna når höga hastigheter finns risken att de kan förloras under acceleration och skada komponenter och arbetare. För att undvika sådana förluster krävs exakt och noggrann kalibrering av var accelererande komponent. Accelerationen sker genom att protonerna låts färdas genom oscillerande elektromagnetiska fält, och för att undvika strålförluster måste dessa fälts amplitud och fas ställas in mycket precist. Denna typ av kalibrering utförs alltid på acceleratoranläggningar, men de etablerade teknikerna inom fältet kan vara mycket tidskrävande, och med en ny accelerator som den på ESS framkommer möjligheten att undersöka nya metoder.

I detta projekt har möjligheten att använda maskininlärning för kalibreringen av accelererande komponenter på ESS undersökts. Maskininlärning är ett samlande begrepp för varierade tekniker, grundade i att ett program utvecklas automatiskt genom att utföra en definierad uppgift, i detta fall en kalibrering, över flera iterationer och uppdatera lösningen med varje iteration. Maskininlärning möjliggör många förbättringar och effektiviseringar av de etablerade teknikerna för protonacceleratorkalibrering. Detta projekt innebär en ny riktning för kombinationen av två viktiga fält i dagens samhälle och forskning, maskininlärning och acceleratorfysik. Om de föreslagna metoderna visar sig effektiva på större skala kan det innebära en ny, starkare standard för ESS och framtida anläggningar. (Less)
Please use this url to cite or link to this publication:
author
Lundquist, Johan LU
supervisor
organization
course
FYSM60 20212
year
type
H1 - Master's Degree (One Year)
subject
keywords
Accelerator, drift tube linac, machine learning, neural network, ESS, linac, tuning, proton, longitudinal dynamics, beam physics
language
English
id
9075906
date added to LUP
2022-02-28 09:23:44
date last changed
2022-02-28 09:23:44
@misc{9075906,
  abstract     = {{The European Spallation Source, currently under construction in Lund, Sweden, will be the world's brightest neutron source. It is driven by a linear accelerator designed to accelerate a beam of protons with 62.5 mA, 2.86 ms long pulses, working at 14 Hz. The final section of its normal-conducting front-end consists of a 39 m long drift tube linac divided into five tanks, designed to accelerate the proton beam from 3.6~MeV to 90 MeV. The high beam current and power impose challenges to the design and tuning of the machine. In order to keep the beam quality throughout the accelerator and beam losses at a minimum the radio frequency amplitude and phase within the accelerating components have to be set within 1% and 1 deg of the design values.

One of the usual methods used to define the radio frequency set-point is called signature matching, which can be a challenging process, and new techniques to meet the growing complexity of accelerator facilities are highly desirable. Machine learning is a rapidly growing field which has found applications in a wide range of scenarios, accelerators being no exception, but the tuning of RF fields using machine learning has yet to be tried. This project explores the possibility of applying machine learning in this area of accelerator physics, comparing this novel technique with the established signature matching and introducing a new possibility for faster tuning using a different data structure. 

For this purpose, simulations of the first tank in the drift tube linac section of the ESS linac were used to produce large amounts of tuning data at varying setpoints for the machine. Data like this was then used with the signature matching and machine learning techniques to fit the necessary functions for the traditional technique and train artificial neural networks for the novel technique. Random machine errors were later introduced to test each method's generalized performance. This data was also restructured to allow for machine tuning in a single shot, while usually a parameter scan is necessary, and machine learning techniques were tried for tuning on this data. 

Machine learning has been found to perform well in comparison with the established method, with some select advantages inherent to machine learning. This faster RF tuning technique only possible with machine learning, is found to perform well, although not quite within the given 1% and 1 deg limitations. As the rest of the results of this project are all on simulated data for the ESS, a comparison with results of these techniques on real data collected from the Spallation Neutron Source in the USA was performed. Future improvements could include workarounds for faulty network inputs and further tuning of the networks used.}},
  author       = {{Lundquist, Johan}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Tuning of the ESS Drift Tube Linac using Machine Learning}},
  year         = {{2022}},
}