Related papers: Time-drift Aware RF Optimization with Machine Lear…
The Proton Improvement Plan phase II (PIP-II) project currently under construction at FNAL will replace the existing 400 MeV normal conducting linac with a new 800 MeV superconducting linac. The beam power in the downstream rapid-cycling…
Spallation source dedicated for neutron scattering experiments, as well as multi-purpose facilities serving several applications call for pulsed mode operation of a high intensity proton linac. There is general agreement on the…
The 40-year-old Fermilab Proton Source machines, constituted by the Pre-Injector, Linac and the synchrotron Booster, have been the workhorse of the Fermi National Accelerator Laboratory (Fermilab). During this time, the High Energy Physics…
Millimeter-wave (mmWave) hybrid analog-digital beamforming is a promising approach to satisfy the low-latency constraint in multiple unmanned aerial vehicles (UAVs) systems, which serve as network infrastructure for flexible deployment.…
The intensity of charged particle beams at Fermilab must be kept within pre-determined safety and operational envelopes in part by assuring all beam within a few percent has been transported from any source to destination. Beam intensity…
The reliability of machine learning (ML) software systems is heavily influenced by changes in data over time. For that reason, ML systems require regular maintenance, typically based on model retraining. However, retraining requires…
The Advanced Superconducting Test Accelerator (ASTA) currently under commissioning at Fermilab will enable a broad range of beam-based experiments to study fundamental limitations to beam intensity and to develop transformative approaches…
PIP-II is the Fermilab's flagship project for providing powerful, high-intensity proton beams to the laboratory's experiments. The heart of PIP-II is an 800-MeV superconducting linac accelerator. It will be located in a new tunnel with new…
The Fermilab accelerator complex delivers intense high-energy proton beams to a variety of fixed-target scientific programs, including a flagship long-baseline neutrino program. With the advent of the Deep Underground Neutrino Experiment…
FAST linear accelerator has been commissioned in 2017. Experimental program of the facility requires high quality beams with well-defined properties. Solenoidal fields at photoinjector, laser spot shape, space charge forces and other…
Resource allocation in integrated sensing and communication (ISAC) systems needs to be optimized to balance the requirements of the communication and sensing modules considering complicated cross-layer data traffic and queue status in…
Beam alignment (BA) is to ensure the transmitter and receiver beams are accurately aligned to establish a reliable communication link in millimeter-wave (mmwave) systems. Existing BA methods search the entire beam space to identify the…
ML-enabled systems that are deployed in a production environment typically suffer from decaying model prediction quality through concept drift, i.e., a gradual change in the statistical characteristics of a certain real-world domain. To…
A superconducting RF accelerator test facility is being constructed at Fermilab. The existing New Muon Lab (NML) building is being converted for this facility. The accelerator will consist of an electron gun, injector, beam acceleration…
Following the PIP-II 800 MeV Linac, Fermilab will need an accelerator that extends from that linac to the MI injection energy of ~8 GeV, completing the modernization of the Fermilab high-intensity accelerator complex. This will maximize the…
The success of reinforcement learning (RL)-based control in tokamaks, an emerging technique for controlled nuclear fusion with improved flexibility, typically requires substantial interaction with a simulator capable of accurately evolving…
The match between the Linac beam energy and the energy determined by the bending field of the Booster magnets is crucial for rf capture, beam quality, and the transmission efficiency in a Booster cycle. The observation of the injection…
Reinforcement Learning (RL) for training Large Language Models is notoriously unstable. While recent studies attribute this to "training inference mismatch stemming" from inconsistent hybrid engines, standard remedies, such as Importance…
In Federated Learning (FL), a number of clients or devices collaborate to train a model without sharing their data. Models are optimized locally at each client and further communicated to a central hub for aggregation. While FL is an…
A room temperature heavy ion linac has been proposed as a new injector of CSRm (the main Cooler Storage Ring) at HIRFL (Heavy Ion Research Facility in Lanzhou), which is expected to improve the performance of HIRFL. The linac injector can…