Results for “Particle beam handling”

DOE OSTI2024

Experimental Safe Extremum Seeking for Accelerators

Here, we demonstrate the recent designs of safe extremum seeking (Safe ES) on the 1-km-long charged particle accelerator at the Los Alamos Neutron Science Center (LANSCE). Safe ES is a modification of extremum seeking (ES) which, in addition to minimizing an analytically unknown cost, also employs a safety filter based on an analytically unknown control barrier function (CBF) safety metric. Tuning is necessitated by accelerators being large complex systems, with many drifting parameters due to thermal effects and degradation. At the same time, safe operation (the maintenance of state constraints) is crucial, as damage brings astronomical costs, both financially and in operation downtime. Our measured (but analytically unknown) safety metric is the beam current. We perform multivariable Safe ES on three accelerator applications, in which we adapt 4, 6, and 3 magnet strength parameters, respectively. Two of the three applications are for validated simulation models of beamlines at LANSCE: the first for the proton radiography (pRad) beamline of 800-MeV protons for spot size tuning; the second on a high-performance code, HPSim, for tuning the low-energy beam transport (LEBT) region that contains a beam of 750-keV protons. The third is an experimental tuning of the steering magnets in the LEBT at LANSCE.

43 PARTICLE ACCELERATORS↗

DOE OSTI2025

Radioisotope production at the Spallation Neutron Source: Design concept of experimental target station

Completion of the Proton Power Upgrade Project for the Spallation Neutron Source (SNS) accelerator at Oak Ridge National Laboratory opens an opportunity to utilize reserve beam power of more than 100 kW for applications beyond neutron production. One of these applications is the production of critical radionuclides. To demonstrate the feasibility of using the reserve beam power to produce radioisotope at SNS, a design concept of a small-scale experimental target station in the Linac Dump area has been developed. This experimental facility will provide isotope yield benchmarking data using protons in the GeV range. It will also enable additional research and development in isotope handling and radiochemical separation. The target station consists of a target module enclosed in a vessel and concrete shielding. Particle transport calculations and thermo-mechanical simulations are used to determine beam parameters, decay time, isotope yield, shielding dimensions, and target design parameters. Calculations verified that the irradiated capsule can be handled manually using hands-off tools and transported to a hot cell in a shielded container for post-irradiation characterizations.

Lee, Yong Joong [ORNL] (ORCID:0000000298381723)↗

DOE OSTI2024

PICOSEC-Micromegas Detector, an innovative solution for Lepton Time Tagging

The PICOSEC-Micromegas (PICOSEC-MM) detector is a novel gaseous detector designed for precise timing resolution in experimental measurements. It eliminates time jitter from charged particles in ionization gaps by using extreme UV Cherenkov light emitted in a crystal, detected by a Micromegas photodetector with an appropriate photocathode. The first single-channel prototype tested in 150 GeV/c muon beams achieved a timing resolution below 25 ps, a significant improvement compared to standard Micropattern Gaseous Detectors (MPGDs). This work explores the specifications for applying these detectors in monitored neutrino beams for the ENUBET Project. Key aspects include exploring resistive technologies, resilient photocathodes, and scalable electronics. New 7-pad resistive detectors are designed to handle the particle flux. In this paper, two potential scenarios are briefly considered: tagging electromagnetic showers with a timing resolution below 30ps in an electromagnetic calorimeter as well as individual particles (mainly muons) with about 20ps respectively.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2026

Continual Learning for Particle Accelerators

Particle accelerators operate under dynamically changing conditions, which often lead to data distribution drifts. These drifts pose significant challenges for Machine Learning (ML) models, which typically fail to maintain performance when faced with such non-stationary data. In particle accelerators, the primary sources of these data drifts include changes in accelerator settings and non-measured parameters such as machine degradation and environmental factors. Previous research has proposed conditional models to handle multiple beam configurations effectively; however, it is challenging to train the ML models on all possible configuration settings. Additionally, conditional models alone can not address performance degradation caused by drifts due to non-measured factors. These limitations contribute to a significant gap between ML development and its deployment in real-world operational settings. To bridge this gap, in this paper, we identify some of the key areas within particle accelerators where continual learning can help mitigate drift-induced performance degradation. In addition, we present a practical use case where a conditional Auto-Encoder model coupled with memory-based continual learning has been employed to demonstrate stable performance even when underlying data drifts.

Schram, Malachi [Thomas Jefferson National Acceler↗

DOE OSTI2026

Continual Learning for Particle Accelerators

Particle accelerators operate under dynamically changing conditions, which often lead to data distribution drifts. These drifts pose significant challenges for Machine Learning (ML) models, which typically fail to maintain performance when faced with such non-stationary data. In particle accelerators, the primary sources of these data drifts include changes in accelerator settings and non-measured parameters such as machine degradation and environmental factors. Previous research has proposed conditional models to handle multiple beam configurations effectively; however, it is challenging to train the ML models on all possible configuration settings. Additionally, conditional models alone can not address performance degradation caused by drifts due to non-measured factors. These limitations contribute to a significant gap between ML development and its deployment in real-world operational settings. To bridge this gap, in this paper, we identify some of the key areas within particle accelerators where continual learning can help mitigate drift-induced performance degradation. In addition, we present a practical use case where a conditional Auto-Encoder model coupled with memory-based continual learning has been employed to demonstrate stable performance even when underlying data drifts

Rajput, Kishansingh [Thomas Jefferson National Acc↗

DOE OSTI2024

Linac_Gen: Integrating Machine Learning and Particle-in-Cell Methods for Enhanced Beam Dynamics at Fermilab

Here, we introduce Linac_Gen, a tool developed at Fermilab, which combines machine learning algorithms with Particle-in-Cell methods to advance beam dynamics in linacs. Linac_Gen employs techniques such as Random Forest, Genetic Algorithms, Support Vector Machines, and Neural Networks, achieving a tenfold increase in speed for phase-space matching in Linacs over traditional methods, through the use of genetic algorithms. Crucially, Linac_Gen's adept handling of 3D field maps elevates the precision and realism in simulating beam instabilities and resonances, marking a key advancement in the field. Benchmarked against established codes, Linac_Gen demonstrates not only improved efficiency and precision in beam dynamics studies but also in the design and optimization of Linac systems, as evidenced in its application to Fermilab's PIP-II Linac project. This work represents a notable advancement in accelerator physics, marrying ML with PIC methods to set new standards for efficiency and accuracy in accelerator design and research. Linac_Gen exemplifies a novel approach in accelerator technology, offering substantial improvements in both theoretical and practical aspects of beam dynamics.

43 PARTICLE ACCELERATORS↗

DOE OSTI2024

Linac_Gen: integrating machine learning and particle-in-cell methods for enhanced beam dynamics at Fermilab

Here, we introduce Linac_Gen, a tool developed at Fermilab, which combines machine learning algorithms with Particle-in-Cell methods to advance beam dynamics in linacs. Linac_Gen employs techniques such as Random Forest, Genetic Algorithms, Support Vector Machines, and Neural Networks, achieving a tenfold increase in speed for phase-space matching in linacs over traditional methods through the use of genetic algorithms. Crucially, Linac_Gen's adept handling of 3D field maps elevates the precision and realism in simulating beam instabilities and resonances, marking a key advancement in the field. Benchmarked against established codes, Linac_Gen demonstrates not only improved efficiency and precision in beam dynamics studies but also in the design and optimization of linac systems, as evidenced in its application to Fermilab's PIP-II linac project. This work represents a notable advancement in accelerator physics, marrying ML with PIC methods to set new standards for efficiency and accuracy in accelerator design and research. Linac_Gen exemplifies a novel approach in accelerator technology, offering substantial improvements in both theoretical and practical aspects of beam dynamics.

43 PARTICLE ACCELERATORS↗

DOE OSTI2024

An Updated Simulation of the Booster Neutrino Beam

For accelerator neutrino experiments, an accurate prediction of the incoming neutrino flux is crucial for reducing uncertainties for all physics measurements. In this exciting period for the Short-Baseline Neutrino program at Fermilab, with far detector (ICARUS) already operating and the near detector (SBND) nearing operation, an updated flux model for the Booster Neutrino Beam (BNB) is presented. The BNB delivers 8 GeV protons to a beryllium target, subsequently producing neutrinos from the decay of the secondary beam of hadrons. A updated Monte-Carlo simulation of beam production in GEANT4 has been created, allowing predictions to be made for detectors with different baselines, offsets and sizes. This new simulation contains new features, such as a full neutrino ancestry to handle hadron production systematics with more precision, with a view to storing all resulting particles - including neutral mesons - from the proton-Beryllium scatter to allow the study of exotic BSM scenarios. Results are presented, with comparisons to the original flux simulated for the MiniBooNE experiment.

Paton, Josephine↗

DOE OSTI2024

MLOps for Beam Controls

Machine learning operations (MLOps) is the standardization and streamlining of the ML development lifecycle to address the challenges associated with large-scale machine learning applications. The full MLOps pipeline consists of open-source tools: DataHub, MinIO and MLflow. It is being used for dataset management and model development to handle changing data dependencies, varying business needs, reproducibility, and diverse teams working with differing tools and skills. To demonstrate the completion of an MLOps pipeline for particle accelerator operations, we are deploying a simple script that computes settings for the Booster’s gradient magnet power supply. Once the demonstration is complete, we will develop and deploy ML-based optimization algorithms to improve Booster’s overall efficiency. This MLOps pipeline opens the gate to systematically develop and deploy ML applications for accelerator controls and diagnostics.

43 PARTICLE ACCELERATORS↗

DOE OSTI2024

MLOps for Beam Controls

Machine learning operations (MLOps) is the standardization and streamlining of the ML development lifecycle to address the challenges associated with large-scale machine learning applications. The full MLOps pipeline consists of open-source tools: DataHub, MinIO and MLflow. It is being used for dataset management and model development to handle changing data dependencies, varying business needs, reproducibility, and diverse teams working with differing tools and skills. To demonstrate the completion of an MLOps pipeline for particle accelerator operations, we are deploying a simple script that computes settings for the Booster’s gradient magnet power supply. Once the demonstration is complete, we will develop and deploy ML-based optimization algorithms to improve Booster’s overall efficiency. This MLOps pipeline opens the gate to systematically develop and deploy ML applications for accelerator controls and diagnostics.

43 PARTICLE ACCELERATORS↗

DOE OSTI2024

MLOps for Beam Controls

Machine learning operations (MLOps) is the standardization and streamlining of the ML development lifecycle to address the challenges associated with large-scale machine learning applications. The full MLOps pipeline consists of open-source tools: DataHub, MinIO and MLflow. It is being used for dataset management and model development to handle changing data dependencies, varying business needs, reproducibility, and diverse teams working with differing tools and skills. To demonstrate the completion of an MLOps pipeline for particle accelerator operations, we are deploying a simple script that computes settings for the Booster’s gradient magnet power supply. Once the demonstration is complete, we will develop and deploy ML-based optimization algorithms to improve Booster’s overall efficiency. This MLOps pipeline opens the gate to systematically develop and deploy ML applications for accelerator controls and diagnostics.

43 PARTICLE ACCELERATORS↗

DOE OSTI2025

Explainable and Differentiable Reinforcement Learning for Multi-objective Optimization in Particle Accelerators

Operating particle accelerators involves optimizing multiple goals simultaneously, which can be challenging due to trade-offs among objectives. While evolutionary algorithms like the genetic algorithm (GA) have been used for various Multi-Objective Optimization (MOO) tasks, they are not inherently suited for complex control problems. This talk highlights two variations of Reinforcement Learning (RL) for concurrently optimizing heat load and trip rates at the Continuous Electron Beam Accelerator Facility (CEBAF). The problem involves strict constraints on individual states, actions, and overall energy requirements of the beam. First, this talk highlights how differentiability can be harnessed through a Deep Differentiable Reinforcement Learning (DDRL) approach to address MOO issues within particle accelerators. We examine the DDRL method alongside Model Free Reinforcement Learning (MFRL), GA, and Bayesian Optimization (BO). The performance of these methods is assessed by generating a Pareto-front for two objectives. Our findings indicate that DDRL excels in handling high-dimensional problems more effectively than MFRL, BO, and GA. Next, we will show integration of explainable physics-based constraints into RL algorithms to enhance trans- parency and trust in decision-making processes by enabling users to verify that agents adhere to established physical principles. This surrogate function can be modeled using neural networks or sparse dictionary mod- els. By examining the mathematical form of the learned constraint function, we are able to confirm the agent has learned to use the established physics of each environment provided but the surrogate model. In addi- tion, we find that the introduction of a mathematical functional dictionary based surrogate model enables our reinforcement learning algorithms to reliably converge for difficult high-dimensional accelerator controls environments.

Rajput, Kishansingh [Thomas Jefferson National Acc↗

DOE OSTI2025

Virtual to Physical: Reinforcement Learning to Optimize SNS Particle Accelerator Controls

Complex accelerators must have control systems that can handle dynamic nonlinear environments. This makes traditional control methods unsuitable as they can struggle to adapt to these uncertainties. This provides an ideal environment for reinforcement learning algorithms as they are adaptable and generalizable. We present a reinforcement learning pipeline that can effectively handle the dynamics of a complex accelerator. We test and prove our pipelines capabilities on multiple environments including the Spallation Neutron Source (SNS) and the Beam Test Facility (BTF) at Oakridge National Lab (ORNL). Due to the limited time available to train an online algorithm like reinforcement learning on a real accelerator, we utilize a virtual twin accelerator (VIRAC) developed by ORNL to pretrain the policy and show its ability to converge in the virtual environment. We then test the adaptability of the pretrained RL model by applying it on the real accelerator and comparing the results. Utilizing our Scientific Optimization and Controls Toolkit (SOCT) and open-source standards such as Gymnasium we create and solve for a MEBT orbit correction problem in the SNS and an emittance maximization problem in the BTF. We show how Twin Delayed Deep Deterministic Policy Gradient (TD3) can solve this optimization environment in the virtual accelerator and transfer this policy onto the real accelerator for inference and model retraining. We show how reinforcement learning can be utilized as a control system for complex accelerators and provide a model pipeline for how an implementation performs and can be adapted to new accelerator control problems.

Kasparian, Armen [Thomas Jefferson National Accele↗

DOE OSTI2024

Achievement in Beam Power Records for the NOvA Target System

We began upgrading the NOvA target system for 1-Mega Watt (1-MW) beam operation in 2017. Major challenges included maintaining the quality of neutrino beams with reliable instrumentation, reducing instantaneous beam heating on the target, increasing cooling power to handle the high-power beam, and controlling tritium water production rate. We finally achieved a one-hour beam power record of 1.018 MW in Summer 2024. This milestone demonstrates our capability to operate at 2+ MW beam power for the future Long Baseline Neutrino Facility (LBNF) and Deep Underground Neutrino Experiment (DUNE).

43 PARTICLE ACCELERATORS↗

DOE OSTI2025

Modelling detector-specific reconstruction uncertainties in LAr-TPC

The Short-Baseline Neutrino (SBN) program features three Liquid Argon Time Projection Chamber (LAr-TPC) detectors positioned along the Booster Neutrino Beam (BNB) axis: the Short Baseline Neutrino Near Detector, MicroBooNE, and the ICARUS T600. As the largest operational LAr-TPC, ICARUS T600 serves as the far detector, located 600 m from the BNB target. While its primary goal is to record neutrino events, it also detects other ionizing events, including cosmic rays. This work focuses on analyzing and modeling detector-specific reconstruction uncertainties in LAr-TPC. These inefficiencies, identified during the Pattern Recognition phase handled by the PANDORA algorithm, impact subsequent Particle Fits and Offline Analysis. Specifically, inaccuracies in track reconstruction can lead to significant physical consequences, such as erroneous particle energy estimates and poor Particle Identification (PID), reducing the efficiency of neutrino event characterization. A key issue addressed is split tracks, caused by missing hits or incomplete track stitching by PANDORA. The aim of this internship is to characterize, model, and quantify the impact of split tracks on track reconstruction.

43 PARTICLE ACCELERATORS↗

DOE OSTI2026

Microsecond-latency feedback at a particle accelerator by online reinforcement learning on hardware

The commissioning and operation of future large-scale scientific experiments will challenge current tuning and control methods. Reinforcement learning (RL) algorithms are a promising solution due to their ability to dynamically adapt to changing environments and consider delayed consequences. In many real-world applications, RL policies must produce actions in real time, often within microseconds to milliseconds, imposing significant constraints on system latency and computational overhead that conventional machine learning libraries are not designed to handle. To control phenomena in real time at these timescales, RL needs to be deployed on-the-edge, namely on dedicated hardware located near the system it controls, without relying on a host CPU or cloud-based inference. In this work we present the design and deployment of an experience accumulator system in a particle accelerator. In this system, deep-RL algorithms run using hardware acceleration and act within a few microseconds, enabling the use of RL for control of phenomena like beam instabilities. The training uses the collected data offline to reduce the number of operations carried out on the acceleration hardware. The proposed architecture was tested in real experimental conditions at the Karlsruhe research accelerator, a synchrotron light source, where the system was used to control artificially induced horizontal betatron oscillations in real-time, with a control loop period of just 2.7 μs. The results showed a performance comparable to the commercial feedback system available at the accelerator, demonstrating the viability and potential of this approach. Due to the self-learning and reconfiguration capability of this implementation, a seamless application to other control problems is possible. Applications range from particle accelerators to large-scale research and industrial facilities.

FPGA↗

DOE OSTI2025

Towards An Updated Simulation of the Booster Neutrino Beam

Precise, accurate neutrino flux predictions for neutrino beam experiments are crucial for physics results. Flux predictions for the Booster Neutrino Beam at Fermilab were first published in 2009 by the MiniBooNE collaboration. It is no longer possible to run the simulation used to create these predictions due to outdated software versions. In this exciting period for the Short-Baseline Neutrino Program at Fermilab, with both the far detector (ICARUS) and the near detector (SBND) operating, an updated flux model for the BNB is necessary. For the purpose of using a dynamic, up-to-date beam model, the SBN program has created a new simulation named G4BNB. This framework contains new features, such as a full neutrino ancestry to handle hadron production systematics with more precision, and the inclusion of neutral mesons from the proton-beryllium scatter to allow the study of exotic BSM scenarios.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

Spallation Neutron Source Proton Power Upgrade (PPU) Project: Lessons Learned for CD-4

The SNS PPU project goals were to design, build, install and test the equipment necessary to double the accelerator power from 1.4 MW to 2.8 MW and to deliver a 2.0 MW qualified target. PPU also included the provision of a stub-out in the SNS accumulator-ring-to-target tunnel to facilitate a rapid connection to a new proton beamline for the Second Target Station (STS) project. The power capability was doubled by increasing the proton beam energy by 33% and the peak beam current by 50%, relative to pre-PPU accelerator performance. The project also included modifications to some buildings and services. The PPU project accomplished the energy upgrade by fabricating and installing new superconducting radiofrequency (RF) cryomodules, with supporting RF equipment, in the existing linac tunnel and klystron gallery, respectively. The High Voltage Converter Modulators (HVCM) and klystrons for some of the existing installed RF equipment were upgraded to handle the higher beam current. The increased beam power of 2 MW on the First Target Station (FTS) was enabled by the addition of a new high-volume gas injection system for pressure pulse and cavitation mitigation in the mercury target and a redesigned mercury target vessel.

43 PARTICLE ACCELERATORS↗