Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “real-time algorithm”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

California Price Response Potential Study

California's energy landscape is undergoing a significant transformation, driven by the increasing integration of renewable energy sources, the increased adoption of distributed energy resources, the electrification of end-use loads, and the growing need for grid efficiency. To address these challenges, recent revisions to the State’s Load Management Standards (LMS) require all of California’s large utilities and community choice aggregators (CCAs) to offer dynamic electricity pricing options to customers by 2027. Dynamic pricing, which involves varying electricity rates based on real-time supply and demand conditions, offers a promising solution for optimizing grid operations, reducing costs, and incentivizing efficient use of grid capacity. Effective implementation of dynamic pricing requires understanding the potential impacts on customer bills, system load, and the cost-effectiveness of automation technologies. This study aims to evaluate the load response of various end-use devices to hourly dynamic prices. The end-uses studied here are space cooling, space heating, water heating, crop irrigation, pool and spa pumps, and electric vehicle (EV) charging, all for both residential and commercial applications, except for crop irrigation. In 2030, these end uses are forecasted to account for 18% of annual electricity demand in the state, but 40% of demand in the peak net load hour. By modeling possible price-responsive load dispatch algorithms and assessing the resulting impacts on both individual bills and the overall grid, we seek to inform policymakers and utilities about the potential benefits and challenges associated with dynamic pricing, and considerations for the design of dynamic pricing tariffs. Additionally, we will explore the cost effectiveness of adopting automation technologies to enable devices to respond more effectively to real-time price signals. This study considers a range of price profiles, accounting for differences across utilities and customer classes, and presents scenarios for dynamic price design via variation in the percentage of total customer electric costs that are allocated dynamically (versus constituting a fixed portion of the hourly volumetric price). We present results focused primarily on 2030, forecasting electricity prices under both low and high-cost scenarios, to inform longer-term tariff design considerations. We design tariffs by starting with 2019 prices that were calculated according to CalFUSE guidance (CPUC, 2022) and that have been used in recent studies; these prices are all-in volumetric rates that vary by utility and are revenue-neutral to each customer class. They are developed by considering six electricity cost components that are allocated hourly based on system load indicators (gross and net load, and wholesale prices). These prices are forecasted to 2030 for low and high cost scenarios, considering recent trends in total electricity costs with and without years of substantial wildfire mitigation investments. These tariffs, which allocate all costs on an hourly basis, are considered our “Full” dynamic tariff design scenario, while two additional scenarios explore allocating a portion of costs as a flat volumetric charge: the “Medium” scenario allocates 50% of revenue dynamically (and keeps 50% flat), while the “Mild” scenario allocates 20% of revenue dynamically. The 20% dynamic allocation on the Mild scenario aims to represent a case where only the marginal operating costs of the grid are included in the dynamic price.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2026

Machine Learning on Heterogeneous, Edge, and Quantum Hardware for Particle Physics (ML-HEQUPP)

The next generation of particle physics experiments will face a new era of challenges in data acquisition, due to unprecedented data rates and volumes along with extreme environments and operational constraints. Harnessing this data for scientific discovery demands real-time inference and decision-making, intelligent data reduction, and efficient processing architectures beyond current capabilities. Crucial to the success of this experimental paradigm are several emerging technologies, such as artificial intelligence and machine learning (AI/ML) and silicon microelectronics, and the advent of quantum algorithms and processing. Their intersection includes areas of research such as low-power and low-latency devices for edge computing, heterogeneous accelerator systems, reconfigurable hardware, novel codesign and synthesis strategies, readout for cryogenic or high-radiation environments, and analog computing. This white paper presents a community-driven vision to identify and prioritize research and development opportunities in hardware-based ML systems and corresponding physics applications, contributing towards a successful transition to the new data frontier of fundamental science.

Gonski, Julia [SLAC]↗
DOE OSTI2026

The high level trigger and express data production at STAR

To meet the demands of the Beam Energy Scan phase-II (BES-II) program, the STAR experiment at the Relativistic Heavy Ion Collider (RHIC) developed a dual real-time framework consisting of a High Level Trigger (HLT) and an Express Data Production system (xProduction). The HLT operates online within the Data Acquisition (DAQ) chain on a dedicated multi-core CPU cluster with the option to offload compute-intensive kernels to Xeon Phi coprocessors. It uses parallelized algorithms, such as the Cellular Automaton (CA) Track Finder, to perform rapid tracking, vertexing, and event filtering. This allows it to select events of interest in real time and provide immediate feedback on detector and beam conditions. In contrast, the xProduction workflow runs concurrently and independently of the DAQ loop. It applies near offline-quality calibration and reconstruction within hours of data collection. The xProduction input is the express data stream, whose content can be enriched by HLT trigger/priority selections under DAQ/HLT resource constraints, and it uses the STAR calibration/conditions framework, incorporating online calibration/QA information when available. This enables early preliminary physics analysis, including the reconstruction of rare signals, such as hyperons and hypernuclei. It also provides collaboration-wide access to analysis-ready datasets. Together, the HLT and xProduction systems form a complementary architecture: the HLT performs online event selection while the xProduction chain delivers high-quality results within a short amount of time. This integrated framework has enabled the prompt reconstruction of the $^5_Λ$ He hypernucleus with high statistical significance and the efficient processing of hundreds of millions of heavy-ion collision events. In conclusion, its demonstrated scalability and robustness establish a model for future high-luminosity experiments requiring both online event filtering and rapid access to analysis-quality data.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2026

Enabling integrated AI control on DIII-D: a control system design with state-of-the-art experiments

We present the design and application of a general algorithm for Prediction And Control using MAchiNe learning (PACMAN) in DIII-D. Machine learning (ML)-based predictors and controllers have shown great promise in achieving regimes in which traditional controllers fail, such as tearing mode (TM) free scenarios, ELM-free scenarios and stable advanced tokamak conditions. The architecture presented here was deployed on DIII-D to facilitate the end-to-end implementation of advanced control experiments, from diagnostic processing to final actuation commands. This paper describes the detailed design of the algorithm and explains the motivation behind each design point. We also describe several successful ML control experiments in DIII-D using this algorithm, including a reinforcement learning controller targeting advanced non-inductive plasmas, a wide-pedestal quiescent H-mode ELM predictor, an Alfvén Eigenmode controller, a Model Predictive Control plasma profile controller and a state-machine TM predictor-controller. There is also discussion on guiding principles for real-time ML controller design and implementation.

machine learning↗
DOE OSTI2026

ClassNMSW- a real-time classification approach for non-recycled municipal solid waste using hyperspectral imaging

Real-time classification of non-recycled municipal solid waste (NMSW) is essential for efficient valorization. This study introduces ClassNMSW, a comprehensive framework for classifying 22 NMSW subclasses under industrial constraints by using hyperspectral imaging (HSI). A primary innovation of this work is the development of a variance-controlled spectral extraction algorithm. Unlike traditional methods that rely on simple averaging, this approach systematically investigates the extent of pixel extraction to minimize the loss of critical chemical information while maximizing data reduction thus ensuring high spectral fidelity with low computational cost. The approach developed in this work integrates automated, computer-vision-based background removal, eliminating the need for the manual thresholding common in current literature. To resolve ambiguities among chemically similar subclasses, a tiered classification and multi-camera fusion strategy (NIR17 and NIR22) is implemented. Results demonstrate that ClassNMSW achieves an object-wise weighted accuracy of 98.70% for single-sensor configurations and 100% under sensor fusion. A novel rolling-window strategy satisfies desired end-to-end latency of <2 s, satisfying the strict deterministic requirements of high-speed industrial sorting environments. The ClassNMSW framework provides a scalable foundation for advancing circularity and resource recovery in large-scale waste valorization operations.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2026

Real-Time Anomaly Detection for Beyond Standard Model Searches in ProtoDUNE Horizontal Drift

This paper summarizes work conducted throughout a SULI internship at Fermi National Accelerator Laboratory focused on building an unsupervised machine learning model for real-time anomaly detection in ProtoDUNE Horizontal Drift. Using simulated data, we trained an autoencoder model on a pure cosmic dataset, and evaluated it on both cosmic and neutrino events---making the model an anomaly detector. The goal was to make a model which matches or exceeds the current ADC Simple Window trigger algorithm so that our model can perform at the same rate but provide sensitivity to potential beyond-the-Standard-Model (BSM) signatures. In the end, we were able to construct a model which slightly exceeds the capabilities of the ADC Simple Window while remaining completely unsupervised, achieving $31.9 \pm 0.2$\% ($26.6 \pm 0.2$\%) $\nu$ efficiency at 5 Hz (2 Hz), a 3.6 (3.2) percentage point increase. Additionally, $17.5 \pm 0.3$\% ($18.3 \pm 0.3$\%) of the events that passed the autoencoder at 5 Hz (2 Hz) were missed by the current trigger algorithm. Future work will investigate alternative normalization methods, including quantile transformation, and evaluate the model on ProtoDUNE-HD detector-glitch data if that data becomes available.

Wilson, Cameron C. [Cincinnati U., RWC]↗
DOE OSTI2026

Real-Time Anomaly Detection for Beyond Standard Model Searches in ProtoDUNE Horizontal Drift

This paper summarizes work conducted throughout a SULI internship at Fermi National Accelerator Laboratory focused on building an unsupervised machine learning model for real-time anomaly detection in ProtoDUNE Horizontal Drift. Using simulated data, we trained an autoencoder model on a pure cosmic dataset, and evaluated it on both cosmic and neutrino events---making the model an anomaly detector. The goal was to make a model which matches or exceeds the current ADC Simple Window trigger algorithm so that our model can perform at the same rate but provide sensitivity to potential beyond-the-Standard-Model (BSM) signatures. In the end, we were able to construct a model which slightly exceeds the capabilities of the ADC Simple Window while remaining completely unsupervised, achieving $31.9 \pm 0.2$\% ($26.6 \pm 0.2$\%) $\nu$ efficiency at 5 Hz (2 Hz), a 3.6 (3.2) percentage point increase. Additionally, $17.5 \pm 0.3$\% ($18.3 \pm 0.3$\%) of the events that passed the autoencoder at 5 Hz (2 Hz) were missed by the current trigger algorithm. Future work will investigate alternative normalization methods, including quantile transformation, and evaluate the model on ProtoDUNE-HD detector-glitch data if that data becomes available.

Wilson, Cameron C. [Cincinnati U., RWC]↗
DOE OSTI2026

Real-Time Anomaly Detection for Searches Beyond the Standard Model in the ProtoDUNE Horizontal Drift Detector

This paper summarizes work conducted throughout a SULI internship at Fermi National Accelerator Laboratory focused on building an unsupervised machine learning model for real-time anomaly detection in ProtoDUNE Horizontal Drift. Using simulated data, we trained an autoencoder model on a pure cosmic dataset, and evaluated it on both cosmic and neutrino events—making the model an anomaly detector. The goal was to make a model which matches or exceeds the current ADC Simple Window trigger algorithm so that our model can perform at the same rate but provide sensitivity to potential beyond-the-Standard-Model (BSM) signatures. In the end, we were able to construct a model which slightly exceeds the capabilities of the ADC Simple Window while remaining completely unsupervised, achieving 31.9 ± 0.2% (26.6 ± 0.2%) ν efficiency at 5 Hz (2 Hz), a 3.6 (3.2) percentage point increase. Additionally, 17.5 ± 0.3% (18.3 ± 0.3%) of the events that passed the autoencoder at 5 Hz (2 Hz) were missed by the current trigger algorithm. Future work will investigate alternative normalization methods, including quantile transformation, and evaluate the model on ProtoDUNE-HD detector-glitch data if that data becomes available.

Wilson, C. [Cincinnati U., RWC]↗
DOE OSTI2026

Quantum Computing in Next-Generation Transportation Optimization

We explore how quantum computing (QC) can advance transportation optimization, with a focus on two high-impact areas: traffic signal control and vehicle electrification with grid integration. As transportation systems grow in complexity, classical optimization methods increasingly struggle to deliver scalable and efficient solutions, particularly for real-time, data-rich environments. This work identifies key challenges within these two domains where QC may offer advantages, particularly in handling combinatorial decision spaces and dynamic constraints. We begin by outlining the limitations of classical approaches for traffic signal control optimization and electric vehicle charging coordination, highlighting where computational limitations arise. Previous quantum formulations are presented and new formulations are proposed to demonstrate how emerging quantum algorithms, including quantum annealing and the Quantum Approximation Optimization Algorithm, could be leveraged to reformulate and address these problems. We also evaluate the suitability of current quantum hardware and discuss recent trends that indicate when QC may become a viable tool for transportation applications. While acknowledging the present limitations of QC technologies, this poster emphasizes the importance of preparing quantum-compatible models today. By reviewing and establishing formulations that align with the strengths of quantum algorithms, researchers and practitioners can better position themselves to take advantage of QC advancements as they occur. This work aims to provide a practical, forward-looking perspective on the near-term potential of quantum computing in transportation optimization.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2026

Large-scale real-time signal processing in physics experiments: the ALICE TPC FPGA pipeline

For LHC Run 3, the ALICE Time Projection Chamber was upgraded to operate in continuous readout mode. Interaction rates of up to 50 kHz in Pb-Pb collisions require real-time processing of more than 3 TB s -1 of raw detector data. This requirement is met by a custom FPGA-based processing pipeline that performs the complete front-end data treatment fully in-stream, including common-mode correction, pedestal subtraction, ion-tail filtering, zero suppression, and dense data packing. A central element of the design is a highly parallel common-mode correction algorithm operating directly on the streaming data. It robustly identifies signal-free readout channels on a time-bin basis and applies pad-dependent scaling to compensate for local variations in capacitive coupling in the GEM readout. In combination with pedestal subtraction and ion-tail filtering, this enables accurate baseline restoration under extreme high-occupancy conditions, preventing signal loss while efficiently suppressing noise prior to zero suppression. The pipeline operates continuously at the full detector bandwidth and reduces the raw input rate of approximately 3 TB s -1 to about 900 GBps for Pb-Pb collisions at the target interaction rate. Overall, it represents a large-scale FPGA-based real-time signal-processing implementation for high-energy physics detector readout.

Digital signal processing (DSP)↗
DOE OSTI2026

PyJMAK: An Open-Source Python Toolkit for Modeling Solid-State Metallurgical Phase Transformations

Accurate prediction of metallurgical phase transformations is an essential basis for autonomous optimization and rapid part qualification. Several methods can be used to estimate the evolution of phase fractions such as JMAK kinetics-based models, phase-field models, thermodynamic models, and data-driven machine learning models. Thermodynamic and phase-field-based methodologies solve multiphysics equations requiring numerous calibration parameters and significant computational resources. As a result, the computation domain is limited to a point or on order of micron-meters. The data-driven models rely on large datasets from experiments and simulations. While the JMAK model only provides information about phase fraction evolution, it can predict this evolution in near real-time using thermal history and thermodynamic data without restriction on the domain. JMAK models have been popularly used by researchers to model phase transformations occuring during additive manufacturing or over arbitrary temperature profiles. Commercial proprietary software such as Abaqus and Ansys or closed-source in-house implementations offer the ability to model JMAK based kinetics to predict phase transformation. However, these software packages are not open-source or freely available for use and development in conjunction with manufacturing machines, sensors, and machine learning algorithms. In addition, the use of the model is restricted by a license token. In contrast, given temperature profiles at multiple points in the domain, this Python-based PyJMAK model can compute phase evolution in parallel due to its stand-alone modular, voxel-based structure, and it can be executed on high-performance computing resources without any license restrictions.

Prabhune, Bhagya [Oak Ridge National Laboratory (O↗
DOE OSTI2026

Real-Time Detection of Charge Jumps in Superconducting Qubits with a Convolutional Neural Network

Ionizing radiation from cosmic rays and gammas can induce discontinuous jumps in the environmental charge of superconducting qubits (charge jumps), causing correlated errors that challenge fault-tolerant quantum computing while simultaneously providing a detection signature for quantum sensing applications. Current detection methods operate offline, introducing latency incompatible with in-the-loop qubit control. In this paper, an online detector of charge jumps for superconducting qubits, based on a dilated causal convolutional neural network (DCCNN) designed for in-the-loop deployment on the Quantum Instrumentation Control Kit (QICK) platform, is presented. The network is trained on synthetic Ramsey tomography scans generated from qubit templates measured at the Northwestern Experimental Underground Site (NEXUS) at Fermilab, and translated to FPGA firmware via hls4ml with ap_fixed$\langle 16,6 \rangle$ quantization, reaching a per-inference latency of $6.19 μ$s on the Zynq UltraScale+ RFSoC ZCU216. At this operating point the DCCNN matches the detection efficiency of the established offline $χ^2$ algorithm ($0.843 \pm 0.022$ vs. $0.866 \pm 0.020$ on $|Δq| \in [0.1, 0.5] e$ at matched false-positive rate), while requiring no per-qubit hyperparameter tuning. This shifts charge-jump detection from a post-hoc diagnostic to a control-loop primitive, enabling adaptive protocols that respond to radiation-induced events in situ, with applications to quantum-computing error mitigation and to the use of superconducting qubits as particle detectors.

Gaytan-Villarreal, Daniel [Carnegie Mellon U.]↗
DOE OSTI2026

A machine learning based approach to online electron reconstruction at CLAS12

Online reconstruction is key for monitoring purposes and real time analysis in High Energy and Nuclear Physics experiments. A necessary component of reconstruction algorithms is particle identification that combines information left by a particle passing through several detector components to identify the particle’s type. Of particular interest to electro-production Nuclear Physics experiments such as CLAS12 is electron identification which is used to trigger data recording. A machine learning approach was developed for CLAS12 to reconstruct and identify electrons by combining raw signals at the data acquisition level from several detector components. Here, this approach achieves an electron identification purity above 75% whilst retaining an efficiency close to 100%. The machine learning tools are capable of running at high rates exceeding the data acquisition rates and will allow electron reconstruction in real-time. This work enhances online analyses and monitoring and can contribute to improved triggering at CLAS12. This machine learning driven approach will also be crucial for experiments aiming to transition to streaming readout operations where online reconstruction will be a key component of the data taking paradigm.

Artificial intelligence↗
DOE OSTI2026

radkit base v1.6

The radkit (base) software suite (python) consists of three primary libraries: stark, trajan, and curie. The trajan library provides the tools to analyze and manipulate data from lidar and inertial measurement unit (IMU) devices, cameras, as well as trajectories from algorithms such as simultaneous localization and mapping (SLAM). These components allow reading and writing standard data formats, performing rigid affine transformations, discretizing three-dimensional space, and visualizing data products. The curie library comprises a standard set of object-oriented tools for radiation data and analysis in the following modules: (1) listmode and binmode data classes with methods for manipulation, plotting, slicing and file IO; (2) radiological/nuclear source detection/identification analysis results; (3) source encounters of correlated analyses and (4) energy-dependent angular detector response functions. The stark package provides low-level tools that are leveraged by both curie and trajan. The tools are flexible for offline analysis as well as performant for real-time integrations.

Salathe, Marco [Lawrence Berkeley National Laborat↗
DOE OSTI2026

Computational capacity in hydrodynamic real-time hybrid simulation applied to simulate the dynamic response of floating offshore wind turbines

Real-time hybrid simulation (RTHS) mitigates similitude distortions in model-scale tests of floating offshore wind turbines (FOWTs) by coupling physical experiments with numerical models in real time. The coupling requires faster-than-real-time numerical computations to satisfy temporal similitude with the physical experiment, presenting a bottleneck for using more complex numerical models in RTHS. This paper presents a hydrodynamic-RTHS (hydro-RTHS) framework for FOWTs that simulates the hydrodynamics physically and the aerodynamics numerically with sensor feedback from the physical testing. The framework adapts the three-loop hardware architecture to leverage greater computational resources and mitigate strict temporal requirements, enabling more computationally demanding numerical analyses in hydro-RTHS. The three-loop hardware architecture integrates multiple machines, each dedicated to either numerical analysis or RTHS controls, with a rate-transition algorithm to synchronize the tasks executed across the different machine processors. Virtual and physical tests verified and validated the hydro-RTHS framework, respectively. The ”virtual” tests, which approximates the physical domain numerically, verified the RTHS framework with respect to a numerical full-scale complete FOWT model simulated in the open-source software, OpenFAST. The virtual tests were able to maintain comparable control signals while enabling greater computational resources for the numerical calculations. Real-world physical tests demonstrated that the hydro-RTHS framework computes aerodynamic forces similar to the complete OpenFAST model, validating the hydro-RTHS framework using the three-loop hardware architecture. Findings show that the hydro-RTHS framework with the three-loop hardware architecture is computationally efficient, with reserve capacity to simulate more complex problems due to the customized software, hardware, and rate-transition algorithm.

17 WIND ENERGY↗
DOE OSTI2026

Trigonometric continuous-variable gates and hybrid quantum simulations of the sine-Gordon model

Hybrid qubit-qumode quantum computing platforms provide a natural setting for simulating interacting bosonic quantum field theories. However, existing continuous-variable gate constructions rely predominantly on polynomial functions of canonical quadratures. In this work, we introduce a complementary universality paradigm based on trigonometric continuous-variable gates, which enable a Fourier-like representation of bosonic operators and are particularly well suited for periodic and non-perturbative interactions. We present an ancilla-based framework for implementing trigonometric gates with arguments given by arbitrary Hermitian functions of qumode quadratures. The protocol yields unitary gates deterministically, and non-unitary gates through probabilistic post-selection. As a concrete application, we develop a hybrid qubit-qumode quantum simulation of the lattice sine-Gordon model. Using these gates, we prepare ground states via quantum imaginary-time evolution, simulate real-time dynamics, compute time-dependent vertex two-point correlation functions, and extract quantum kink profiles under topological boundary conditions. Our results demonstrate that trigonometric continuous-variable gates provide a physically natural framework for simulating interacting field theories on near-term hybrid quantum hardware, while establishing a parallel route to universality beyond polynomial gate constructions. We expect that the trigonometric gates introduced here to find broader applications, including quantum simulations of condensed matter systems, quantum chemistry, and biological models.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2026

Optimization-based approaches to control of connected and automated vehicles: Principles, complexities, applications, challenges, and outlook

Safe and optimal motion control for connected and automated vehicles (CAVs) poses a fundamental optimization challenge at the intersection of system complexity, environmental uncertainty, and stringent real-time constraints. Existing surveys address this challenge in isolation – focusing either on specific control techniques or individual uncertainty sources – without providing a unified framework that characterizes the trade-offs among computational tractability, performance verifiability, and adaptive generalization across paradigms. This review addresses that gap by presenting a cohesive analytical framework concentrated on the decision-making and trajectory optimization layers of the CAV autonomy stack. We systematically analyze three major optimization paradigms – first-principles model-based optimization, data-driven methods, and hybrid synergistic architectures – evaluating each against four core complexity axes: problem formulation, constraint handling, optimality guarantees, and robustness. Key applications including platooning, trajectory planning, collision avoidance, and cooperative control are examined to reveal recurring methodological patterns and critical operational constraints that limit real-world performance. Our synthesis identifies verifiable hybrid architectures, incentive-aligned multi-agent cooperation, and hardware-algorithm co-design as the defining research frontiers, and distills a targeted agenda for developing CAV control systems that are simultaneously safe, computationally efficient, and deployable in the full complexity of real-world traffic environments.

Muzahid, Abu Jafar Md [University of Tennessee, Kn↗
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↗