Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “network robustness”

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 1,009 records · Page 56

DDStore: Distributed Data Store for Scalable Training of Graph Neural Networks on Large Atomistic Modeling Datasets

Graph neural networks (GNNs) are a class of Deep Learning models used in designing atomistic materials for effective screening of large chemical spaces. To ensure robust prediction, GNN models must be trained on large volumes of atomistic data on leadership class supercomputers. Even with the advent of modern architectures that consist of multiple storage layers that include node-local NVMe devices in addition to device memory for caching large datasets, extreme-scale model training faces I/O challenges at scale.We present DDStore, an in-memory distributed data store designed for GNN training on large-scale graph data. DDStore provides a hierarchical, distributed, data caching technique that combines data chunking, replication, low-latency random access, and high throughput communication. DDStore achieves near-linear scaling for training a GNN model using up to 1000 GPUs on the Summit and Perlmutter supercomputers, and reaches up to a 6.15x reduction in GNN training time compared to state-of-the-art methodologies.

Choi, Jong Youl↗

Omics-Driven Biotechnology for Industrial Applications

Biomanufacturing is a key component of biotechnology that uses biological systems to produce bioproducts of commercial relevance, which are of great interest to the energy, material, pharmaceutical, food, and agriculture industries. Biotechnology-based approaches, such as synthetic biology and metabolic engineering are heavily reliant on “omics” driven systems biology to characterize and understand metabolic networks. Knowledge gained from systems biology experiments aid the development of synthetic biology tools and the advancement of metabolic engineering studies toward establishing robust industrial biomanufacturing platforms. In this review, we discuss recent advances in “omics” technologies, compare the pros and cons of the different “omics” technologies, and discuss the necessary requirements for carrying out multi-omics experiments. We highlight the influence of “omics” technologies on the production of biofuels and bioproducts by metabolic engineering. Finally, we discuss the application of “omics” technologies to agricultural and food biotechnology, and review the impact of “omics” on current COVID-19 research.

59 BASIC BIOLOGICAL SCIENCES↗

Dependable classical-quantum computing systems engineering

Increasing evidence suggests quantum computing (QC) complements traditional High-Performance Computing (HPC) by leveraging its unique capabilities, leading to the emergence of a new, hybrid paradigm, QHPC. However, this integration introduces new challenges, with dependability–defined by reproducibility, resiliency, and security and privacy–emerging as a central concern for building trustworthy systems that provide an advantage to the users. This paper proposes a framework for dependable QHPC system design, organized around these three pillars. We identify integration challenges, anticipate roadblocks, and highlight productive synergies across QC, HPC, cloud platforms, and network security. Drawing from both classical computing principles and quantum-specific insights, we present a roadmap for co-design that supports robust hybrid architectures. Our approach offers concrete metrics for assessing dependability, provides design guidance for engineers working at the QC-HPC interface, and surfaces new engineering questions around complexity, scale, and fault tolerance. Ultimately, designing for dependability is key to realizing practical, scalable QHPC systems and accelerating the broader quantum ecosystem capable of translating quantum promises into actual application delivery.

HPC↗

Scalable deep learning for watershed model calibration

Watershed models such as the Soil and Water Assessment Tool (SWAT) consist of high-dimensional physical and empirical parameters. These parameters often need to be estimated/calibrated through inverse modeling to produce reliable predictions on hydrological fluxes and states. Existing parameter estimation methods can be time consuming, inefficient, and computationally expensive for high-dimensional problems. In this paper, we present an accurate and robust method to calibrate the SWAT model (i.e., 20 parameters) using scalable deep learning (DL). We developed inverse models based on convolutional neural networks (CNN) to assimilate observed streamflow data and estimate the SWAT model parameters. Scalable hyperparameter tuning is performed using high-performance computing resources to identify the top 50 optimal neural network architectures. We used ensemble SWAT simulations to train, validate, and test the CNN models. We estimated the parameters of the SWAT model using observed streamflow data and assessed the impact of measurement errors on SWAT model calibration. We tested and validated the proposed scalable DL methodology on the American River Watershed, located in the Pacific Northwest-based Yakima River basin. Our results show that the CNN-based calibration is better than two popular parameter estimation methods (i.e., the generalized likelihood uncertainty estimation [GLUE] and the dynamically dimensioned search [DDS], which is a global optimization algorithm). For the set of parameters that are sensitive to the observations, our proposed method yields narrower ranges than the GLUE method but broader ranges than values produced using the DDS method within the sampling range even under high relative observational errors. The SWAT model calibration performance using the CNNs, GLUE, and DDS methods are compared using R 2 and a set of efficiency metrics, including Nash-Sutcliffe, logarithmic Nash-Sutcliffe, Kling-Gupta, modified Kling-Gupta, and non-parametric Kling-Gupta scores, computed on the observed and simulated watershed responses. The best CNN-based calibrated set has scores of 0.71, 0.75, 0.85, 0.85, 0.86, and 0.91. The best DDS-based calibrated set has scores of 0.62, 0.69, 0.8, 0.77, 0.79, and 0.82. The best GLUE-based calibrated set has scores of 0.56, 0.58, 0.71, 0.7, 0.71, and 0.8. The scores above show that the CNN-based calibration leads to more accurate low and high streamflow predictions than the GLUE and DDS sets. Our research demonstrates that the proposed method has high potential to improve our current practice in calibrating large-scale integrated hydrologic models.

54 ENVIRONMENTAL SCIENCES↗

Hyperaccumulation of Gadolinium by Methylorubrum extorquens AM1 Reveals Impacts of Lanthanides on Cellular Processes Beyond Methylotrophy

Lanthanides (Ln) are a new group of life metals, and many questions remain regarding how they are acquired and used in biology. Methylotrophic bacteria can acquire, transport, biomineralize, and use Ln as part of a cofactor complex with pyrroloquinoline quinone (PQQ) in alcohol dehydrogenases. For most methylotrophic bacteria use is restricted to the light Ln, which range from lanthanum to samarium (atomic numbers 57–62). Understanding how the cell differentiates between light and heavy Ln, and the impacts of these metals on the metabolic network, will advance the field of Ln biochemistry and give insights into enzyme catalysis, stress homeostasis, and metal biomineralization and compartmentalization. We report robust methanol growth with the heavy Ln gadolinium by a genetic variant of the model methylotrophic bacterium Methylorubrum extorquens AM1, named evo -HLn, for “ evo lved for H eavy L antha n ides.” A non-synonymous single nucleotide polymorphism in a cytosolic hybrid histidine kinase/response regulator allowed for sweeping transcriptional alterations to heavy metal stress response, methanol oxidation, and central metabolism. Increased expression of genes for Ln acquisition and uptake, production of the Ln-chelating lanthanophore, PQQ biosynthesis, and phosphate transport and metabolism resulted in gadolinium hyperaccumulation of 36-fold with a trade-off for light Ln accumulation. Gadolinium was hyperaccumulated in an enlarged acidocalcisome-like compartment. This is the first evidence of a bacterial intracellular Ln-containing compartment that we name the “lanthasome.” Carotenoid and toblerol biosynthesis were also upregulated. Due to its unique capabilities, evo -HLn can be used to further magnetic resonance imaging (MRI) and bioremediation technologies. In this regard, we show that gadolinium hyperaccumulation was sufficient to produce MRI contrast in whole cells, and that evo -HLn was able to readily acquire the metal from the MRI contrast agent gadopentetic acid. Finally, hyperaccumulation of gadolinium, differential uptake of light and heavy Ln, increased PQQ levels, and phosphate transport provide new insights into strategies for Ln recovery.

59 BASIC BIOLOGICAL SCIENCES↗

The suitability of differentiable, physics-informed machine learning hydrologic models for ungauged regions and climate change impact assessment

As a genre of physics-informed machine learning, differentiable process-based hydrologic models (abbreviated as δ or delta models) with regionalized deep-network-based parameterization pipelines were recently shown to provide daily streamflow prediction performance closely approaching that of state-of-the-art long short-term memory (LSTM) deep networks. Meanwhile, δ models provide a full suite of diagnostic physical variables and guaranteed mass conservation. Here, we ran experiments to test (1) their ability to extrapolate to regions far from streamflow gauges and (2) their ability to make credible predictions of long-term (decadal-scale) change trends. We evaluated the models based on daily hydrograph metrics (Nash–Sutcliffe model efficiency coefficient, etc.) and predicted decadal streamflow trends. For prediction in ungauged basins (PUB; randomly sampled ungauged basins representing spatial interpolation), δ models either approached or surpassed the performance of LSTM in daily hydrograph metrics, depending on the meteorological forcing data used. They presented a comparable trend performance to LSTM for annual mean flow and high flow but worse trends for low flow. For prediction in ungauged regions (PUR; regional holdout test representing spatial extrapolation in a highly data-sparse scenario), δ models surpassed LSTM in daily hydrograph metrics, and their advantages in mean and high flow trends became prominent. In addition, an untrained variable, evapotranspiration, retained good seasonality even for extrapolated cases. The δ models' deep-network-based parameterization pipeline produced parameter fields that maintain remarkably stable spatial patterns even in highly data-scarce scenarios, which explains their robustness. Combined with their interpretability and ability to assimilate multi-source observations, the δ models are strong candidates for regional and global-scale hydrologic simulations and climate change impact assessment.

54 ENVIRONMENTAL SCIENCES↗

A real time neural net estimator of fatigue life

A neural net architecture is proposed to estimate, in real-time, the fatigue life of mechanical components, as part of the Intelligent Control System for Reusable Rocket Engines. Arbitrary component loading values were used as input to train a two hidden-layer feedforward neural net to estimate component fatigue damage. The ability of the net to learn, based on a local strain approach, the mapping between load sequence and fatigue damage has been demonstrated for a uniaxial specimen. Because of its demonstrated performance, the neural computation may be extended to complex cases where the loads are biaxial or triaxial, and the geometry of the component is complex (e.g., turbopump blades). The generality of the approach is such that load/damage mappings can be directly extracted from experimental data without requiring any knowledge of the stress/strain profile of the component. In addition, the parallel network architecture allows real-time life calculations even for high frequency vibrations. Owing to its distributed nature, the neural implementation will be robust and reliable, enabling its use in hostile environments such as rocket engines. This neural net estimator of fatigue life is seen as the enabling technology to achieve component life prognosis, and therefore would be an important part of life extending control for reusable rocket engines.

Troudet, T.↗

Decentralized Formation Flying Control in a Multiple-Team Hierarchy

This paper presents the prototype of a system that addresses these objectives-a decentralized guidance and control system that is distributed across spacecraft using a multiple-team framework. The objective is to divide large clusters into teams of manageable size, so that the communication and computational demands driven by N decentralized units are related to the number of satellites in a team rather than the entire cluster. The system is designed to provide a high-level of autonomy, to support clusters with large numbers of satellites, to enable the number of spacecraft in the cluster to change post-launch, and to provide for on-orbit software modification. The distributed guidance and control system will be implemented in an object-oriented style using MANTA (Messaging Architecture for Networking and Threaded Applications). In this architecture, tasks may be remotely added, removed or replaced post-launch to increase mission flexibility and robustness. This built-in adaptability will allow software modifications to be made on-orbit in a robust manner. The prototype system, which is implemented in MATLAB, emulates the object-oriented and message-passing features of the MANTA software. In this paper, the multiple-team organization of the cluster is described, and the modular software architecture is presented. The relative dynamics in eccentric reference orbits is reviewed, and families of periodic, relative trajectories are identified, expressed as sets of static geometric parameters. The guidance law design is presented, and an example reconfiguration scenario is used to illustrate the distributed process of assigning geometric goals to the cluster. Next, a decentralized maneuver planning approach is presented that utilizes linear-programming methods to enact reconfiguration and coarse formation keeping maneuvers. Finally, a method for performing online collision avoidance is discussed, and an example is provided to gauge its performance.

Mueller, Joseph .↗

On-line fission products measurements during a PWR severe accident: the French DECA-PF project

Following the Fukushima accident, a lot of recommendations was drawn by international organizations (IAEA, OECD, NUGENIA network...) in order to improve the safety in such accidental conditions and mitigate their consequences. One of these recommendations was to improve the robustness of the instrumentation, which was dramatically lacking at Fukushima, as well as to better determine the Source Term involved in nuclear accident. The DECA-PF project (Diagnosis of a degraded reactor core through Fission Product measurements) was elaborated in this context and selected as one of 21 collaborative R and D projects in the field of nuclear safety and radioprotection, funded in May 2013 by the French National Research Agency. Over the months following the Fukushima accident, a CEA crisis team was held in order to analyze on-line the situation taking into account the data delivered by TEPCO and other organizations. Despite the difficulties encountered concerning the reliability of these data, the work performed showed the high capacity of Fission Products (FP) measurements to get a diagnosis relative to the status of the reactors and the spent fuel pools (SFP). Based on these FP measurements, it was possible to conclude that the main origin of the releases was coming from the cores and not from the SFP, in particular for SFP-4 which was of high concern, and that the degradation level of the reactors was very large, including probably an extensive core melting. To improve the reliability of this kind of diagnosis, the necessity to get such measurements as soon as possible after the accident and as near as possible from the reactor was stressed. In this way the present DECA-PF project intends to develop a new and innovative instrumentation taking into account the design of the French nuclear power plants on which sand bed filters have been implemented for severe accident management. Three complementary techniques, devoted to measure the FP release on-line, are being studied: - Gamma spectrometry, with an industrial objective to build a prototype aimed at improving the capacity of the present radiation monitoring system, - Gas chromatography, for the quantification of the fission gases (Xe, Kr) as well as potential carbon oxides produced in case of Molten Corium Concrete Interaction, - Optical absorption spectroscopy, the objective of this most innovative technique being to quantify the tetra-oxide of ruthenium, which could be produced in case of lower head failure, and the gaseous forms of iodine (molecular and organic) released in the environment. A global description and the present status of this project is presented, focusing on the Source Term establishment at the outlet stack of the sand bed filters and on the perspectives of implementation of the on-line gamma spectrometry equipment. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Enhancing Unknown Waveform Detection by Learning Intra and Inter-domain Dependencies with Advanced Attention Fusion Mechanisms

Detection of unknown waveforms in mission-critical communications is a crucial area of interest for the Department of Energy (DoE). Traditional methods and recent deep learning-based approaches often assume that the training set includes all possible classes, which is impractical for detecting new waveforms. This limitation gives rise to the problem of open-set recognition (OSR), which involves correctly identifying known classes while detecting and rejecting unknown or unseen classes. To address this limitation, we propose a novel dual-domain complex-valued neural architecture that jointly processes time-domain and frequency-domain signal representations using transformer mechanisms. A transformer model is a deep learning architecture that uses self-attention mechanisms to process and learn relationships in sequential data. Our model employs a cosine similarity loss to extract domain-specific features and incorporates a transformer architecture in the latent space to weigh the importance of different features from the time and frequency domains. The transformer layer includes stacked self-attention and cross-attention modules to learn intra-domain and inter-domain dependencies, creating a more holistic signal representation. An attention-based fusion module intelligently combines the time and frequency-domain features using multi-head attention, enabling the network to learn the optimal feature for each domain in each input signal. Quantitative results demonstrate the impact of these architectural choices on overall performance, showing significant improvement after incorporating self and cross-attention modules and using complex attention fusion over simple weighted fusion. Our ongoing work will focus on addressing the limitations of threshold-based OSR methods by developing a novel generative framework that integrates a conditional diffusion probabilistic model (DPM). DPM is a generative framework that learns to synthesize complex data by reversing a gradual noising process using a neural network trained to denoise step-by-step. Our goal is to leverage the inherent strengths of DPMs for identifying unknown signals more robustly. One primary advantage of using a DPM is its ability to provide a more reliable anomaly score based on the model's reconstruction error, rather than relying solely on classifier confidence. Additionally, the iterative denoising process of DPMs makes this approach naturally resilient to low Signal-to-Noise Ratio (SNR) conditions, where traditional methods often fail. By implementing this generative framework, we aim to enhance the model's capability to accurately detect unknown waveforms and maintain performance in challenging environments.

99 - GENERAL AND MISCELLANEOUS↗

Large Scale Bilevel Optimization for N-K SCOPF Using Adversarial Robustness

Ensuring a secure dispatch against multiple simultaneous outages has long been desired to maintain grid security in the presence of severe events, such as extreme weather phenomena. Traditionally denoted as N-k security constrained optimal power flow (N-k SCOPF), this problem is intractable to solve due to its size being combinatorial in the number of simultaneous outages and due to the non-convex nature of the AC network constraints. This hinders the use of N-k SCOPF for operating realistic-scale systems. In this paper, we introduce a methodology to scalably solve an AC-feasible dispatch that improves security over k simultaneous outages. Our methodology poses N-k SCOPF as a bilevel optimization problem and solves it using an adversarial robustness approach. We develop new efficient methods to solve each level of the bilevel optimization by employing knowledge of the physics of the underlying system. This yields significant improvements in speed and convergence that enable us to address the N-k SCOPF problem at scale. We demonstrate the effectiveness of our method by conducting a comprehensive analysis of an N-3 SCOPF for a 500-bus network. Furthermore, we emphasize the ability of our physics-driven techniques to handle larger systems by successfully scaling up to 12,000 buses.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Explainable AI (XAI)-driven vibration sensing scheme for surface quality monitoring in a smart surface grinding process

Local Interpretable and Model-agnostic Explanation (LIME), an explainable artificial intelligence (XAI) approach is adapted to identify the globally important time-frequency bands for predicting average surface roughness (Ra) in a smart grinding process. The smart grinding setup consisted of a Supertech CNC precision surface grinding machine, instrumented with a Dytran piezoelectric accelerometer attached to the tailstock quill along the tangential direction (Y-axis). For every grinding pass, vibration signatures were captured, and the ground truth surface roughness values were recorded using a Mahr Marsurf M300C portable surface roughness profilometer. The roughness values ranged from 0.06 to 0.14 microns over the complete set of experiments. Time-frequency domain spectrogram frames were extracted for each of the vibration signals collected during the grinding process. Convolutional Neural Networks (CNNs) were modeled to predict the surface roughness based on these spectrogram frames and their image augmentations. The best CNN model was able to predict the roughness values with an overall R2-score of 0.95, training R2-score of 0.99, and testing R2-score of 0.81 with only 80 sets of vibration signals corresponding to 4 experiments with 20 trials each. Although the data size is not large enough to guarantee such performance metrics in real-world scenarios, one can extract statistically consistent explanations underlying the relationships these complex deep learning models capture. Further, the LIME methodology was implemented on the developed surface roughness CNN model to identify the important time-frequency bands (i.e., the superpixels of a spectrogram) influencing the predictions. Based on the identified important regions on the spectrogram frames, the corresponding frequency characteristics were determined that influence the surface roughness predictions. The important frequency range based on LIME results was approximately 11.7 to 19.1 kHz. The power of XAI was demonstrated by cutting down the sampling rate from 160 kHz to 30, 20, 10, and 5 kHz based on the important frequency range and considering Nyquist criteria. Separate CNN models were developed for these ranges by only extracting time-frequency contents below their corresponding Nyquist cut-offs. A proper data acquisition strategy is proposed by comparing the model performances to argue the selection of a sufficient sampling rate to capture the grinding process successfully and robustly.

42 ENGINEERING↗

CO 2 storage site characterization using ensemble-based approaches with deep generative models

Estimating spatially distributed properties such as permeability from available sparse measurements is a great challenge in efficient subsurface CO 2 storage operations. In this paper, a deep generative model that can accurately capture complex subsurface structure is tested with an ensemble-based inversion method for accurate and accelerated characterization of CO 2 storage sites. We chose Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) for its realistic reservoir property representation and Ensemble Smoother with Multiple Data Assimilation (ES-MDA) for its robust data fitting and uncertainty quantification capability. WGAN-GP are trained to generate high-dimensional permeability fields from a low-dimensional latent space and ES-MDA then updates the latent variables by assimilating available measurements. Several subsurface site characterization examples including Gaussian, channelized, and fractured reservoirs are used to evaluate the accuracy and computational efficiency of the proposed method and the main features of the unknown permeability fields are characterized accurately with reliable uncertainty quantification. Furthermore, the estimation performance is compared with a widely-used variational, i.e., optimization-based, inversion approach, and the proposed approach outperforms the variational inversion method in several benchmark cases. We explain such superior performance by visualizing the objective function in the latent space: because of nonlinear and aggressive dimension reduction via generative modeling, the objective function surface becomes extremely complex while the ensemble approximation can smooth out the multi-modal surface during the minimization. This suggests that the ensemble-based approach works well over the variational approach when combined with deep generative models at the cost of forward model runs unless convergence-ensuring modifications are implemented in the variational inversion.

42 ENGINEERING↗

Entry, Descent, and Landing Analysis of Low Mass Martian Probe for Surface Climate Network

Aeolus is a mission to provide the first direct measurements of Martian atmospheric wind speeds and correlate them with thermal and compositional data to bring together a complete systematic description for the global energy balance and climate cycles of Mars. Objective A of the Mars Exploration Program Analysis Group (MEPAG) Goal II cites a Science Investigation Area to develop a network of surface landers to provide global, diurnal and synoptic coverage of the near-surface environment to characterize present Mars climate [1]. Aeolus will characterize the present Martian global climate system via an orbiter element and a novel surface network element. A series of low mass, low power probes will be distributed across the Martian surface and measure pressure, sky opacity, temperature, wind speed, water vapor and other trace gases. Low mass and power electrical systems designed to operate in extreme cold and daily thermal cycling for one Martian year could facilitate lower cost, Class-D science probes. Aeolus is targeting a total per-probe mass of <5 kg. Recent advances in nanofabrication of chip-scale sensors that are highly robust to temperature and shock, combined with a novel packaging concept (the deployable sensor probe [2]) originally developed by the Aerospace Corp., for the first time enables a Mars surface climate sensor network. These probes are folded flat during transit to Mars and automatically deploy into the final reentry configuration once released from the orbiter. The capability to fold for stowage significantly increases the probe quantity carried by the orbiter.

EDL↗

Networked Gamma Radiation Detection System for Rapid Tactical Response

A networked gamma radiation detection system with directional sensitivity and energy spectral data acquisition capability is being developed by the National Security Technologies, LLC, to support the close and intense tactical engagement of law enforcement who carry out counterterrorism missions. In the proposed design, three clusters of 2' × 4' × 16' sodium iodide crystals (4 each) with digiBASE-E (for list mode data collection) would be placed on the passenger side of a minivan. To enhance localization and facilitate rapid identification of isotopes, advanced smart realtime localization and radioisotope identification algorithms like WAVRAD (wavelet-assisted variance reduction for anomaly detection) and NSCRAD (nuisance-rejection spectral comparison ratio anomaly detection) will be incorporated. We are testing a collection of algorithms and analysis on the problem of radiation detection with a distributed sensor network, examining such a network’s associated characteristics, and also the trade-offs between false positive alarm rates, true positive alarm rates, and time to detect multiple radiation sources in a large area. Empirical and simulation analyses of critical system parameters, such as number of sensors, sensor placement, and sensor response functions, are being examined. This networked system will provide an integrated radiation detection architecture and framework with (i) a large nationally recognized search database equivalent that would help generate a common operational picture in a major radiological crisis; (ii) a robust reach back connectivity for search data to be evaluated by home teams; and, finally, (iii) a possibility of integrating search data from multi-agency responders.

false alarm rate↗

Temporal Dynamics of Aerodynamic Canopy Height Derived From Eddy Covariance Momentum Flux Data Across North American Flux Networks

Aerodynamic canopy height ( h a ) is the effective height of vegetation canopy for its influence on atmospheric fluxes and is a key parameter of surface-atmosphere coupling. However, methods to estimate h a from data are limited. This synthesis evaluates the applicability and robustness of the calculation of h a from eddy covariance momentum-flux data. At 69 forest sites, annual h a robustly predicted site-to-site and year-to-year differences in canopy heights ( R 2 = 0.88, 111 site-years). At 23 cropland/grassland sites, weekly h a successfully captured the dynamics of vegetation canopies over growing seasons ( R 2 > 0.70 in 74 site-years). Our results demonstrate the potential of flux-derived h a determination for tracking the seasonal, interannual, and/or decadal dynamics of vegetation canopies including growth, harvest, land use change, and disturbance. The large-scale and time-varying h a derived from flux networks worldwide provides a new benchmark for regional and global Earth system models and satellite remote sensing of canopy structure.

54 ENVIRONMENTAL SCIENCES↗

Electron beam energy stabilization using a neural network hybrid controller at the Australian Synchrotron Linac.

This paper describes the implementation of a neural network hybrid controller for energy stabilization at the Australian Synchrotron Linac. The structure of the controller consists of a neural network (NNET) feed forward control, augmented by a conventional Proportional-Integral (PI) feedback controller to ensure stability of the system. The system is provided with past states of the machine in order to predict its future state, and therefore apply appropriate feed forward control. The NNET is able to cancel multiple frequency jitter in real-time. When it is not performing optimally due to jitter changes, the system can successfully be augmented by the PI controller to attenuate the remaining perturbations. With a view to control the energy and bunch length at the FERMI{at}Elettra Free Electron Laser (FEL), the present study considers a neural network hybrid feed forward-feedback type of control to rectify limitations related to feedback systems, such as poor response for high jitter frequencies or limited bandwidth, while ensuring robustness of control. The Australian Synchrotron Linac is equipped with a beam position monitor (BPM), that was provided by Sincrotrone Trieste from a former transport line thus allowing energy measurements and energy control experiments. The present study will consequently focus on correcting energy jitter induced by variations in klystron phase and voltage.

43 PARTICLE ACCELERATORS↗

Integrating 5G Technology for Improved Process Monitoring and Network Slicing in ICS

Industrial Control Systems (ICS) are crucial for monitoring physical processes that support essential cyber-enabled services like power generation. The use of proprietary communication and lack of effective intrusion detection mechanisms pose constraints for efficient operation. Therefore, there is a need to modernize these systems with decentralized technologies like Edge Computing and 5G. However, integrating 5G and Edge Computing into large-scale ICS networks presents implementation and performance challenges. To address these challenges, this paper proposes an integrated ICS architecture that combines 5G and Edge Computing technologies with traditional ICS protocols. The objective is to minimize implementation and operational difficulties while improving the monitoring of physical processes and enabling robust intrusion detection. The proposed architecture outlines the necessary components, services, and communication protocols required for the integration of 5G and Edge Computing.

Aguayo, Jared M.↗