IPAC2021 - List of Authors (Kafkes, D.L.)

Paper	Title	Page
TUPAB327	Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster	2268
	D.L. Kafkes Fermilab, Batavia, Illinois, USA M. Schram JLab, Newport News, Virginia, USA
	Funding: This research was sponsored by the Fermilab Laboratory Directed Research and Development Program under Project ID FNAL-LDRD-2019-027: Accelerator Control with Artificial Intelligence. We describe the offline machine learning (ML) development for an effort to precisely regulate the Gradient Magnet Power Supply (GMPS) at the Fermilab Booster accelerator complex via a Field-Programmable Gate Array (FPGA). As part of this effort, we created a digital twin of the Booster-GMPS control system by training a Long Short-Term Memory (LSTM) to capture its full dynamics. We outline the path we took to carefully validate our digital twin before deploying it as a reinforcement learning (RL) environment. Additionally, we demonstrate the use of a Deep Q-Network (DQN) policy model with the capability to regulate the GMPS against realistic time-varying perturbations.
DOI •	reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB327
About •	paper received ※ 18 May 2021 paper accepted ※ 22 June 2021 issue date ※ 20 August 2021
Export •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

Paper

Title

Page

Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster

2268

D.L. Kafkes
Fermilab, Batavia, Illinois, USA
M. Schram
JLab, Newport News, Virginia, USA

Funding: This research was sponsored by the Fermilab Laboratory Directed Research and Development Program under Project ID FNAL-LDRD-2019-027: Accelerator Control with Artificial Intelligence.
We describe the offline machine learning (ML) development for an effort to precisely regulate the Gradient Magnet Power Supply (GMPS) at the Fermilab Booster accelerator complex via a Field-Programmable Gate Array (FPGA). As part of this effort, we created a digital twin of the Booster-GMPS control system by training a Long Short-Term Memory (LSTM) to capture its full dynamics. We outline the path we took to carefully validate our digital twin before deploying it as a reinforcement learning (RL) environment. Additionally, we demonstrate the use of a Deep Q-Network (DQN) policy model with the capability to regulate the GMPS against realistic time-varying perturbations.

DOI •

reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB327

About •

paper received ※ 18 May 2021 paper accepted ※ 22 June 2021 issue date ※ 20 August 2021

Export •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)