Publications
Publication Information
| Title | Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster |
| Authors | D. Kafkes, Malachi Schram |
| JLAB number | JLAB-CST-21-3502 |
| LANL number | (None) |
| Other number | DOE/OR/23177-5351 |
| Document Type(s) | (Meeting) |
| Associated with EIC: | No |
| Supported by Jefferson Lab LDRD Funding: | No |
| Funding Source: | Nuclear Physics (NP) |
| Other Funding: | FNAL-LDRD-2019-027: Accelerator Control with Artif |
|
Meeting Paper compiled for IPAC21 Proceedings Proceedings of IPAC 2021 Edited By Liu Lin, John M. Byrd, Regis Neuenschwander, Renan Picoreti, Volker R. W. Schaa JACOW (2021) Page(s) 2268-2271 | |
| Publication Abstract: | 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. |
| Experiment Numbers: | |
| Group: | Data Science |
| Document: | |
| DOI: | https://doi.org/10.18429/JACoW-IPAC2021-TUPAB327 |
| Accepted Manuscript: | tupab327.pdf |
| Supporting Documents: | |
| Supporting Datasets: | |