Control theory is a pivotal field of study that focuses on the behavior of dynamical systems and the development of strategies to influence these systems towards desired outcomes. The principle of control theory find its application in plenty of disciplines including engineering, economics, biology and beyond. It were control concepts like the Kalman filter that has flew the Apollo to the...
Multi-objective reinforcement learning (MORL) extends traditional reinforcement learning (RL) by addressing environments where multiple conflicting objectives must be optimized simultaneously. In real-world applications, such as autonomous systems, particle accelerator optimization and control, agents often face trade-offs between competing goals. This lecture provides an overview of the key...
Many accelerator physics problems, such as beamline design, beam dynamics model calibration, online tuning and phase space measurements rely on solving high-dimensional optimisation problems over beam dynamics simulations. Numerical optimisers have successfully been applied to such tasks, but they struggle as the dimensionality and complexity of the objective function increase. In machine...
Large-scale facilities like European XFEL consist of a multitude of subsystems, which often require frequent calibration. Additionally, accurate real-time tuning of many of these subsystems is critical to maintain stable and optimal performance. Automation techniques can be leveraged to reduce operators' time investment and potentially increase the exploitation of allotted beamtime, both in...
Particle accelerators play a critical role in modern scientific research. However, existing manual beam control methods heavily rely on experienced operators, leading to significant time consumption and potential challenges in managing next-generation accelerators characterized by higher beam current and stronger nonlinear properties. In this paper, we establish a dynamical foundation for...
A general Bayesian optimisation tool is being developed at Diamond Light Source to improve machine performance by constructing surrogates from Gaussian Process (GP) models. Priors are placed on covariance kernel hyperparameters to guide an optimiser and prevent overfitting. The model has been integrated with the machine control system. During an experiment aimed at improving injection...
The Linear IFMIF Prototype Accelerator (LIPAc) is designed to accelerate 125 mA of D+ to 9 MeV in CW. The very high power stored in the beam (~1.1 MW) and the use of superconductive RF cavities requires precise control of beam losses (target <1e-6). On the other hand the intense beam is affected by strong space charge forces that easily results in significant halo formation. This contribute is...
This paper presents a novel application of Multipoint Bayesian Algorithmic Execution (multipointBAX) to optimize dynamic aperture (DA) and momentum aperture (MA) in lattice design. DAMA optimization is a critical design task for storage rings, ultimately determining the flux of x-ray sources and luminosity of colliders. Traditionally, solving this multi-objective optimization problem has...
Particle accelerator operation requires simultaneous optimization of
multiple objectives. Multi-Objective Optimization (MOO) is particularly challenging
due to trade-offs between the objectives. Evolutionary algorithms, such as genetic
algorithm (GA), have been leveraged for many optimization problems, however, they
do not apply to complex control problems by design. This paper...
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...
We present a reinforcement learning (RL) framework for controlling particle accelerated experiments that builds explainable physics-based constraints on agent behavior. The goal is to increase transparency and trust by letting users verify that the agent's decision-making process incorporates suitable physics. Our algorithm uses a learnable surrogate function for physical observables, such as...
The National Synchrotron Radiation Centre SOLARIS is a third generation light source. SOLARIS, as a big science facility with seven fully operational beamlines, is obligated to provide the best possible conditions for conducting research. One of the ways to create favorable environment is delivering precise tools for teams working across many different fields in SOLARIS. The general problem...
Plans for the Electron-Ion Collider (EIC), to be built at Brookhaven National Laboratory, include end-to-end and bottom-up capabilities in artificial intelligence (AI) and machine learning (ML). Enabling these capabilities, especially for EIC Operations, will require the large-scale integration of software platforms and tools for the reliable and efficient management of AI/ML-related data,...
Coherent synchrotron radiation (CSR) is a limiting effect in linear accelerators with dispersive elements due to its contribution to projected transverse emittance growth. This effect becomes a limitation for highly compressed beams. Even though CSR-induced projected emittance growth has been widely studied, conventional measurement techniques are not detailed enough to resolve the...
Kicker magnets are essential for particle beam injection and extraction within CERN’s accelerator complex, where high reliability is crucial to maintaining the availability needed for numerous scientific experiments. This study proposes a machine learning approach for forecasting anomalies in these systems, aiming to proactively identify and isolate potential faults before failure occurs. To...
At the European XFEL, detecting anomalies in superconducting cavities is essential for reliable accelerator performance. We began with a model-based fault detection approach focused on residual analysis to identify anomalies. To improve fault discrimination, particularly for quench events, we augmented this system with machine learning (ML) models. Key challenges included the scarcity of...
The vast amount of data generated by accelerators makes manual monitoring impractical due to its labor-intensive nature. Existing machine learning solutions often rely on labeled data, manual inspection, and hyperparameter tuning, which limits their scalability. To address these challenges, we leverage coincidence learning—an unsupervised technique designed for multi-modal tasks—to...
This work presents a machine learning-based approach for compensating magnetic hysteresis in the main dipole and quadrupole magnets of the multi-cycling CERN SPS, utilizing time series neural architectures like the Temporal Fusion Transformers trained on magnetic field measurements. The predicted magnetic fields enable feed-forward, cycle-by-cycle, compensation through the CERN accelerator...
Fast simulations of intense relativistic electron beams can be sufficiently accurate to allow for tuning of an accelerator’s magnetic transport field, but are incapable of capturing all relevant beam physics due to limitations in the model. Because methods that do capture these effects are significantly more computationally-expensive, e.g. particle in-cell simulations, they are fundamentally...
A major challenge in constructing future nuclear fusion power plants is understanding how reactor materials are damaged by the neutron flux generated during the fusion process. In order to address this challenge, the IFMIF-DONES neutron source is being built for material irradiation, generating the necessary neutron flux through a stripping reaction between accelerated deuterons and a lithium...
A multifaceted virtual accelerator model that seamlessly integrates with the online experimental system would highly benefit the operators to test and evaluate beam tuning scenarios and apply them online. As part of this effort, the beam dynamics code TRACK is wrapped with control system architecture and the graphic user interface BADGER developed by SLAC. Customizability and task...
Multimodal Large Language models extend LLMs’ capabilities to input beyond text, often images. At the European XFEL, these models are used as Retrieval-Augmented Generative (RAG) Knowledge assistants in technical and administrative domains. We present a selection of current applications and prototypes: chatbot assistants for data service support, business travel aid, vision-based document...
In modern synchrotron light sources, maintaining beam stability is critical for ensuring high-quality synchrotron radiation performance. Light source stability is governed by stability of current, beam position and beam size. Beam size stability on the order of several microns need to be improved for future experiments. Reinforcement learning (RL) offers a promising approach for real-time beam...
The beam for CERN's North Area proton physics program is produced through a Multi-Turn Extraction (MTE) scheme at the Proton Synchrotron (PS). Using fourth-order resonant excitation, the beam is split into five beamlets in horizontal phase space, with extraction occurring over five consecutive turns. The quality of the splitting is measured by the uniformity of intensities across the beamlets....
Beams typically do not travel through the magnet centers because of errors in storage rings. The beam deviating from the quadrupole centers is affected by additional dipole fields due to magnetic field feed-down. Beam-based alignment (BBA) is often performed to determine a golden orbit where the beam circulates around the quadrupole center axes. For storage
rings with many quadrupoles, the...
Many accelerator physics problems, such as beamline design, beam dynamics model calibration, online tuning and phase space measurements rely on solving high-dimensional optimisation problems over beam dynamics simulations. Numerical optimisers have successfully been applied to such tasks, but they struggle as the dimensionality and complexity of the objective function increase. In machine...
An ML-based optimizer has been working on maximizing XFEL performance at SACLA [1]. The spectral brightness of XFEL was successfully optimized by using a new high-resolution inline spectrometer [2]. To improve the XFEL performance further, we plan to enhance beam diagnostics and simulation environments for the ML-based optimizer. As for beam diagnostics, we are developing an X-band RF...
The APPLE-II types of elliptically polarized undulators (EPU) are extensively utilized in synchrotron light sources. Manufacturing imperfections in the EPU inevitably lead to the creation of a residual skew quadrupole component, which couples horizontal betatron oscillation and dispersion to the vertical plane, consequently altering the vertical beam size. To regulate the vertical beam size,...
The National Synchrotron Radiation Center SOLARIS, third generation light source, is the only synchrotron located in Central-Eastern Europe, in Poland. The SOLARIS Center, with seven fully operational beamlines, serves as a hub for research across a diverse range of disciplines. The most important aspect of such research infrastructure is to provide stable working conditions for the users,...
The Brookhaven Pre-injector Accelerator Facility, which serves RHIC, NSRL, BLIP, and the future EIC, requires occasional tuning of its transfer beam line optics by control room operators to optimize parameters like beam current and emittance. Machine learning (ML) can significantly speed up this tuning process by helping operators quickly identify optimal settings. To facilitate this, ML...
Planned upgrades of the European X-Ray Free Electron Laser (Eu-
XFEL) target higher photon energy and a high duty-cycle operation up to CW-
operation using a superconducting RF gun with lower gradient. An operation in
this regime though critically depends on improvements of the beam slice emit-
tance of the electron gun. Within the OPAL-FEL project, we are addressing this
challenge by...
Superconducting radio frequency (SRF) cavities are critical components in particle accelerators, where accurately calibrated RF signals are essential for assessing cavity bandwidth and detuning, providing key insights into cavity performance and facilitating optimal accelerator operation. In practice, however, calibration drift due to humidity and temperature fluctuations over time poses a...
In 2021, the Chinese ADS Front-end demo superconducting radio-frequency (SRF) linac, known as CAFe, successfully conducted a commissioning of a 10 mA, 200 kW continuous wave proton beam. During this commissioning, it was observed that the SRF faults are the leading causes of short machine downtime trips, contributing to approximately 70% of total beam trips. Analyzing fault data and...
In the SPS, a flexible machine serving the LHC and a multitude of fixed-target experiments and fast-extraction facilities, reliable monitoring of the transverse beam position across a wide range of different beam structures and intensities is essential for stable and efficient operation. Today, the calibration procedure and signal processing of the beam position monitors (BPMs) of the SPS –...
Aging of the stripper foil and unexpected machine shutdowns are the primary causes for reduction of the injected intensity from CERN’s linac3 into the Low Energy Ion Ring (LEIR). As a result, the set of optimal control parameters that maximizes beam intensity in the ring tends to drift, requiring daily adjustments to the machine control settings. In this paper, several data-driven methods such...
The Beam Synchrotron Radiation Longitudinal density monitor (BSRL) at the LHC leverages time-correlated single-photon counting to provide high-dynamic-range measurements of particle populations within each bunch in the LHC including monitoring of “ghost” and “satellite” bunches, which represent charge captured in nominally empty buckets, thereby enhancing the accuracy of luminosity...
Machine learning methods provide a significant potential for the optimized operation of complex facilities, such as particle accelerators. In this contribution, the first training and application of surrogate models to the electron accelerator S-DALINAC based on Fully-Connected Neural Networks (FCNN) will be presented.
An exhaustive data-mining algorithm has been developed to generate the...
Beam tuning in particle accelerators is a complex task, especially when physical modeling is impractical due to the lack of complete beam diagnostics. Traditional methods often rely on iterative manual tuning by operators, which can be inefficient. Reinforcement learning (RL) algorithms offer a promising alternative for automating this process. In this work, we demonstrate the successful...
The Proton Synchrotron (PS) at CERN is equipped with numerous RF systems allowing for evolved longitudinal beam manipulations to adapt the number of bunches and their spacing. The beam produced for the LHC undergoes several bunch splittings, merging and batch compression. Each manipulation must be carefully adjusted to minimize the spread in bunch parameters at PS extraction. The design of...
Particle accelerators, such as the CERN Linear Electron Accelerator for Research (CLEAR), play a critical role in various scientific fields.
Ensuring their operation is automatic, stable, and reproducible is vital for the scalability of future large-scale accelerator projects.
This paper presents an initial step toward autonomous control of the CLEAR beamline, beginning with a basic beam...
