Brian Schupbach
YOU?
Author Swipe
View article: AI-Enabled Operations at Fermi Complex: Multivariate Time Series Prediction for Outage Prediction and Diagnosis
AI-Enabled Operations at Fermi Complex: Multivariate Time Series Prediction for Outage Prediction and Diagnosis Open
The Main Control Room of the Fermilab accelerator complex continuously gathers extensive time-series data from thousands of sensors monitoring the beam. However, unplanned events such as trips or voltage fluctuations often result in beam o…
View article: Beyond PID Controllers: PPO with Neuralized PID Policy for Proton Beam Intensity Control in Mu2e
Beyond PID Controllers: PPO with Neuralized PID Policy for Proton Beam Intensity Control in Mu2e Open
We introduce a novel Proximal Policy Optimization (PPO) algorithm aimed at addressing the challenge of maintaining a uniform proton beam intensity delivery in the Muon to Electron Conversion Experiment (Mu2e) at Fermi National Accelerator …
View article: ML-based Real-Time Control at the Edge: An Approach Using hls4ml
ML-based Real-Time Control at the Edge: An Approach Using hls4ml Open
This study focuses on implementing a real-time control system for a particle accelerator facility that performs high energy physics experiments. A critical operating parameter in this facility is beam loss, which is the fraction of particl…
View article: Edge AI for accelerator controls (READS): beam loss deblending
Edge AI for accelerator controls (READS): beam loss deblending Open
model may be producible to help de-blend losses between machines. Work is underway as part of the Fermilab Real-time Edge AI for Distributed Systems Project (READS) to develop a ML empowered system that collects streamed BLM data and addit…
View article: FPGA Architectures for Distributed ML Systems for Real-Time Beam Loss De-Blending
FPGA Architectures for Distributed ML Systems for Real-Time Beam Loss De-Blending Open
existing operations-critical functions of the BLM system. This paper will focus on the evolution of the architectures, which provided the high-frequency, low-latency collection of synchronized data streams to make real-time inferences. The…
View article: Disentangling Beam Losses in The Fermilab Main Injector Enclosure Using Real-Time Edge AI
Disentangling Beam Losses in The Fermilab Main Injector Enclosure Using Real-Time Edge AI Open
The Fermilab Main Injector enclosure houses two accelerators, the Main Injector and Recycler Ring. During normal operation, high intensity proton beams exist simultaneously in both. The two accelerators share the same beam loss monitors (B…
View article: Artificial Intelligence for improved facilities operation in the FNAL LINAC
Artificial Intelligence for improved facilities operation in the FNAL LINAC Open
The energy consumption in accelerator structures during beam downtimes is a significant fraction of the overall energy budget. Accurate prediction of downtime duration could inform actions to reduce this energy consumption. The LCAPE proje…
View article: Synchronous High-frequency Distributed Readout For Edge Processing At The Fermilab Main Injector And Recycler
Synchronous High-frequency Distributed Readout For Edge Processing At The Fermilab Main Injector And Recycler Open
The Main Injector (MI) was commissioned using data acquisition systems developed for the Fermilab Main Ring in the 1980s. New VME-based instrumentation was commissioned in 2006 for beam loss monitors (BLM)[2], which provided a more systema…
View article: Synchronous High-Frequency Distributed Readout for Edge Processing at the Fermilab Main Injector and Recycler [Poster]
Synchronous High-Frequency Distributed Readout for Edge Processing at the Fermilab Main Injector and Recycler [Poster] Open
VME backplane and broadcast BLM measurements over Ethernet, while not disrupting the existing operations critical functions of the BLM system. This paper will detail the design, implementation, and testing of this parallel data pathway.
View article: FERMILAB PIP-II BOOSTER LLRF UPGRADE
FERMILAB PIP-II BOOSTER LLRF UPGRADE Open
A proposed upgrade to the Fermilab Booster synchrotron is presented. A new LLRF system is required for interfac-ing to the PIP-II Linac that is currently under construction. The present feedback system has legacy analog hardware interleave…
View article: Hydrogen and Cesium Monitor for H<sup>-</sup> Magnetron Sources
Hydrogen and Cesium Monitor for H<sup>-</sup> Magnetron Sources Open
The relative concentration of cesium to hydrogen in the plasma of a H- magnetron source is an important parameter for reliable operations. If there is too much cesium, the surfaces of the source become contaminated with it and sparking occ…
View article: Measuring the Relative Performance of LINAC Beam Outages with L-CAPE Anomaly Detection (Gaussian Mixture Methods &amp; Hierarchical Agglomerative Clustering) [Slides]
Measuring the Relative Performance of LINAC Beam Outages with L-CAPE Anomaly Detection (Gaussian Mixture Methods & Hierarchical Agglomerative Clustering) [Slides] Open
and output cavities for all 24 beams, and individual gain cavities for each cluster. A closely related optional configuration, also for a 10 MW tube, would involve four totally independent cavity clusters with four independent input caviti…
View article: Semantic Regression for Disentangling Beam Losses in the Fermilab Main Injector and Recycler
Semantic Regression for Disentangling Beam Losses in the Fermilab Main Injector and Recycler Open
Fermilab’s Main Injector enclosure houses two accelerators: the Main Injector (MI) and the Recycler (RR). In periods of joint operation, when both machines contain high intensity beam, radiative beam losses from MI and RR overlap on the en…
View article: The L-CAPE Project at FNAL
The L-CAPE Project at FNAL Open
The controls system at FNAL records data asynchronously from several thousand Linac devices at their respective cadences, ranging from 15Hz down to once per minute. In case of downtimes, current operations are mostly reactive, investigatin…
View article: Recent Improvements in the Beam Capture at Fermilab Booster for High Intensity Operation
Recent Improvements in the Beam Capture at Fermilab Booster for High Intensity Operation Open
The Fermilab Booster uses multi-turn beam injection with all its cavities phased such that beam sees a net zero RF voltage even when each station is at the same maxi-mum voltage. During beam capture the RF voltage is increased slowly by us…
View article: Real-time artificial intelligence for accelerator control: A study at the Fermilab Booster
Real-time artificial intelligence for accelerator control: A study at the Fermilab Booster Open
We describe a method for precisely regulating the gradient magnet power supply (GMPS) at the Fermilab Booster accelerator complex using a neural network trained via reinforcement learning. We demonstrate preliminary results by training a s…
View article: Accelerator Real-time Edge AI for Distributed Systems (READS) Proposal
Accelerator Real-time Edge AI for Distributed Systems (READS) Proposal Open
Our objective will be to integrate ML into Fermilab accelerator operations and furthermore provide an accessible framework which can also be used by a broad range of other accelerator systems with dynamic tuning needs. We will develop of r…
View article: Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster
Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster Open
tune measures the tunes in two planes over the energy ramping cycle with an accuracy of 0.01 in real time.
View article: Real-Time Edge AI for Distributed Systems (READS): Progress on Beam Loss De-Blending for the Fermilab Main Injector and Recycler
Real-Time Edge AI for Distributed Systems (READS): Progress on Beam Loss De-Blending for the Fermilab Main Injector and Recycler Open
The Fermilab Main Injector enclosure houses two accelerators, the Main Injector and Recycler. During normal operation, high intensity proton beams exist simultaneously in both. The two accelerators share the same beam loss monitors (BLM) a…
View article: Optimizing Mu2e Spill Regulation System Algorithms
Optimizing Mu2e Spill Regulation System Algorithms Open
A slow extraction system is being developed for the Fermilab’s Delivery Ring to deliver protons to the Mu2e experiment. During the extraction, the beam on target experiences small intensity variations owing to many factors. Various adaptiv…
View article: Real-time Artificial Intelligence for Accelerator Control: A Study at\n the Fermilab Booster
Real-time Artificial Intelligence for Accelerator Control: A Study at\n the Fermilab Booster Open
We describe a method for precisely regulating the gradient magnet power\nsupply at the Fermilab Booster accelerator complex using a neural network\ntrained via reinforcement learning. We demonstrate preliminary results by\ntraining a surro…
View article: MEBT Laser Notcher (Chopper) for Booster Loss Reduction
MEBT Laser Notcher (Chopper) for Booster Loss Reduction Open
The Fermilab Booster, which utilizes multi-turn injection and adiabatic capture, the extraction gap (aka "notch") has been created in the ring at injection energy using fast kickers which deposit the beam in a shielded absorber within the …