Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Optimization of reliability management”

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 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Real-Time Automated pH Control within Batch Processes Relying on Raman pH Measurement

Nuclear fission is an energy source that can provide consistent power with very low associated carbon emissions. However, management of the used nuclear fuel is an important aspect of the application of nuclear power. Recycling of useful components from used fuel is an attractive option, but this involves chemical processing of the fuel. Possible chemical separation technologies that might be used in this regard are sensitive to solution pH. Raman spectroscopy is a promising technique for monitoring the pH of solutions in real time. Classical pH probes are too fragile to be used in the harsh environments encountered in nuclear fuel processing. Raman probes are robust and can withstand these harsh environments to track pH. Coupled with chemometric analysis, the demonstration of the use of Raman spectroscopy to track and predict the pH in carboxylate-buffered systems is made possible. Utilizing this spectroscopy in conjunction with Programmable Logic Controllers mimics industrial control systems used in many modern industrial settings. This showcases a pragmatic approach toward leveraging Raman spectroscopy and chemometric model outputs as inputs for a real-time control system. The model to predict pH created by chemometrics proved to be successful in tracking pH. The optimal pH for TALSPEAK extraction of lanthanides and actinides from aqueous solution is known to proceed in a narrow pH range of around pH = 2.8 ± 0.1. This study uses Raman optical monitoring and automated control to return and maintain solution pH within this range after acid or base perturbations move the solution pH well outside this region. Root-mean-square errors show that pH changes measured using Raman spectroscopy on the batch process solution are reliably measured and used to automatically correct and maintain solution pH. Measurement of solution pH tracks favorably with electrochemical pH probe comparison measurements. As a result, the ability to showcase Raman spectroscopy paired with chemometrics analysis acts as a durable, better alternative data source compared to traditional pH probes to optimize the separation efficiency in the used nuclear fuel processing.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Assessment of Technoeconomic Opportunities in Automation for Nuclear Microreactors

Achieving full decarbonization of all economic sectors remains a challenge, especially in niche markets. For example, remote communities and industrial or mining activities detached from the main electric grid heavily rely on fossil fuels, similar to urban and industrial microgrids with combined heat and power needs. A combination of renewables and energy storage is often not suitable due to cost, reliability, intermittency, and large storage requirements. Small nuclear reactors with a flexible purpose could serve these applications. Microreactors (MR) are a class of reactors that are compact, factory manufactured, transportable, and self-regulating. Typically, they generate much less power than their large reactor counterparts. The main advantages of microreactors include the versatile nature of the energy produced, the reliability of supply, and freedom from having to transport and store large quantities of fuels on-site, coupled with the absence of dependence on an electrical grid. A strong business case is needed to move from the microreactor prototype to the commercialization phase. In fact, fossil fuels are still relatively inexpensive, and in the near term, carbon credits will be available to virtually compensate for emissions. For microreactors, one of the main costs in operation and maintenance (O&M) is their staffing levels. In this study, we investigate how to optimize the number (and thus the cost) of workers, moving from a traditional, fully manned, on-site personnel approach to an unmanned, remote personnel approach. We examine four different staffing models that can be implemented as the technology matures and evolves. We estimate the staffing needs of each model and build a business case to justify the substitution of on-site personnel with adequate technologies. To do so, we propose a cost model to quantify potential cost reductions from automating O&M activities. The model accounts for both the reduction in cost derived from the reduced number of full-time-equivalent (FTE) employees and the increase in cost derived from the need to buy new control hardware as needed. Applying the cost model that we created to different scenarios, an on-site O&M cost reduction exceeding 80% can be expected. Additionally, we found that it is more impactful to focus on automating routine O&M tasks rather than attempting to automate transient management (shutdowns, restarts, monitoring condition deviations). In fact, transients typically account for less than 1% of the total FTE time spent on the reactors.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2024

Bayesian batch optimization for molybdenum versus tungsten inertial confinement fusion double shell target design

Access to reliable, clean energy sources is a major concern for national security. Much research is focused on the “grand challenge” of producing energy via controlled fusion reactions in a laboratory setting. For fusion experiments, specifically inertial confinement fusion (ICF), to produce sufficient energy, the fusion reactions in the ICF fuel need to become self-sustaining and burn deuterium-tritium (DT) fuel efficiently. The recent record-breaking NIF ignition shot was able to achieve this goal as well as produce more energy than used to drive the experiment. This achievement brings self-sustaining fusion-based power systems closer than ever before, capable of providing humans with access to secure, renewable energy. In order to further progress toward the actualization of such power systems, more ICF experiments need to be conducted at large laser facilities such as the United States's National Ignition Facility (NIF) or France's Laser Mega-Joule. The high cost per shot and limited number of shots that are possible per year make it prohibitive to perform large numbers of experiments. As such, experimental design relies heavily on complex predictive physics simulations for high-fidelity “preshot” analysis. These multidimensional, multi-physics, high-fidelity simulations have to account for a variety of input parameters as well as modeling the extreme conditions (pressures and densities) present at ignition. Such simulations (especially in 3D) can become computationally prohibitive to turn around for each ICF experiment. In this work, we explore using Bayesian optimization with Gaussian processes (GPs) to find optimal designs for ICF double shell targets, while keeping computational costs to manageable levels. These double shell targets have an inner shell that grades from beryllium on the outer surface to the higher Z material molybdenum, as opposed to the nominally used tungsten, on the inside in order to trade off between the high performance associated with high density inner shells and capsule stability. We describe our results for “capsule-only” xRAGE simulations to study the physics between different capsule designs, inner shell materials, and potential for future experiments.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Gearbox bearing crack growth prognostics and uncertainty quantification with physics-informed machine learning

This paper introduces the extreme theory of functional connections (X-TFC), a physics-informed machine learning algorithm, and tailors it to estimate the remaining useful life (RUL) of wind turbine gearbox bearings experiencing fatigue crack growth. Unlike purely data-driven methods, X-TFC embeds a physics model, based on Head's theory in this work, into its training objective. The core of X-TFC is a random-projection single-layer neural network trained via an extreme learning machine, which requires only limited damage progression data and solves for output weights with a least-squares optimization algorithm. A composite loss function balances the network's fit to observed degradation data against the residuals of the governing crack growth differential equation, ensuring the learned damage trajectory remains physically plausible. When applied to a vibration-based health-index (HI) dataset measured during the growth of a crack on the inner ring of a high-speed bearing in a wind turbine gearbox (Bechhoefer and Dubé, 2020), X-TFC achieves near-zero prediction bias. Even when trained on only the first 10 %–20 % of the damage progression data, with sufficient physics weighting its predictions remain monotonic and smooth, delivering high prognosability and trendability. To quantify the epistemic uncertainty, we employ a Monte Carlo ensemble of independently initialized X-TFC models trained on noise-perturbed data, which yields confidence intervals around each RUL estimate and captures both model-parameter and epistemic uncertainty. In addition to a vibration-based HI, we demonstrate that the proposed framework can be directly applied to a supervisory control and data acquisition (SCADA) data-based HI (Eftekhari Milani et al., 2026) measured during similar wind turbine gearbox bearing crack faults, preserving its accuracy and interpretability. This extension shows the versatility of our approach, which is applicable to bearings of multiple gearbox manufacturers, models, and ratings using only SCADA data. By integrating domain knowledge with machine learning, X-TFC offers a rapid, reliable tool for crack prognostics. Its adaptability to other bearing failure modes, such as pitch bearing ring cracks, positions X-TFC as a powerful enabler of data-driven, physics-informed asset management in the wind energy sector and beyond.

17 WIND ENERGY↗
DOE OSTI2021

Igiugig Site Visit Report

The National Renewable Energy Laboratory (NREL) team conducted a site visit during January 23-25, 2019 with the Igiugig Village Council (IVC) and other stakeholders to assist the community of Igiugig in refining their long-term energy strategy. The agenda included a tour of the village, data collection, presentations, long-term visioning exercises, identification of energy scenarios, and a school presentation. Participants in the site visit activities spanned a range of local, regional, government, and industry participation. There were 24 attendees at the all-day community workshop on Thursday, January 24th, with participants from the IVC, NREL, Bristol Bay Native Association (BBNA), Bristol Bay Native Corporation (BBNC), Lake and Peninsula Borough, Alaska Energy Authority, Intergrid, ORPC, Deer Stone Consulting, the Southwest Alaska Municipal Conference, and University of Alaska Fairbanks - Alaska Center for Energy and Power. Igiugig is a well-organized rural Alaskan community, with a leadership team that appears to have broad community support. The use of consensus-based decision making is one tangible example of how the leadership team actively invokes and promotes community-focused thinking. The NREL team observed anecdotal evidence of how this approach seems to be strengthening the community: active participation in the visioning exercise, a student hosted fund-raiser dinner at the school, a clean and organized landfill, and well-maintained roads and buildings. Igiugig has wind, solar, and river hydrokinetic resources readily available within the community. Wind has shown to be a promising resource in the region, and has been integrated into microgrids around the state, but the community has had mixed success with wind technologies: some devices failed quickly and others continue to operate. The economics of solar energy are improving in Alaska, and economical solar projects are being installed around the state. Considering that economic activity in Igiugig peaks during the summer sport fishing season, solar could prove to be a valuable supplement to the electrical system. Igiugig has been a test site for two different river hydrokinetic devices, and they have began a third project to operate ORPC's RivGen device, which is delivering valuable device performance data, operations and maintenance experience, and design refinement information. Already, this device has made over 7-million revolutions, and delivered over 8 MWh of power to the community. Igiugig's river resource is fairly unique because it is available year-round and has the potential to provide reliable base-load power for months at a time. A preliminary investigation of this diverse resource mix suggests that Igiugig could achieve very high levels (70% or more) of annual renewable generation contributions. A critical step in pursuing this path is identifying the mix of energy assets (generation and storage) that best meets the community's budget, needs, and goals. The assets that a community installs early in their grid-modernization initiative can constrain the options that are economical at later stages, which may lead to sub-optimal solutions. This is where technical and economic analysis of potential scenarios (i.e., distinct mixes of energy assets) can be useful in identifying the most promising pathways so that a community can make informed and strategic decisions about the assets they install. These analyses are most accurate and informative when they are based on actual technology performance and cost data. As the RivGen project continues to operate and generate this data, we will be better prepared to evaluate the technology's long-term viability and to identify research areas that would improve it. Igiugig's energy projects are at the cutting edge of two intersecting technology areas: 1) deploying an operational river hydrokinetic turbine, and 2) integrating renewable energy sources to achieve very high percent renewables. If these projects are successful, the lessons learned and technologies developed could be valuable for other microgrids around the world. Igiugig's unique resource mix make it an ideal location for this work, and the community's organizational strength make it an ideal partner in this ground-breaking work. Ongoing support for technical assistance to manage and address technical challenges along the way will maximize the probability of project success.

13 HYDRO ENERGY↗
DOE OSTI2026

Toward an AI-Powered Software Pipeline for Real-Time Tracking and Analysis of Wildfire and Smoke

Real-time tracking of wildfires and smoke is crucial for effective response, minimizing damage, protecting lives, and efficiently managing resources during fire emergencies. We develop a web-based AI-powered pipeline that detects wildfires in aerial video and estimates deployment-relevant behavior metrics, including cumulative burned area, burned-area growth rate, fire spread direction, and smoke dispersion. The system combines a YOLO-based detector with YCbCr-based fire segmentation, HSV-based smoke segmentation, Farneback optical flow, and centroid-based spatiotemporal tracking. Using ground sampling distance (GSD), pixel-level fire masks are converted to physical burned-area measurements by correlating fire pixel counts with camera altitude and tilt angle. We benchmark YOLO variants and non-YOLO baselines (GoogLeNet, CNN, DBN, Autoencoder, U-Net, and AlexNet) on the IEEE FLAME dataset and a newly created aerial frame dataset, Wildfire-DB. Cross-dataset evaluation uses a strict threshold-transfer protocol: decision thresholds are selected on FLAME validation and transferred unchanged to Wildfire-DB to quantify generalization under domain shift. YOLOv6 achieves the strongest cross-dataset frame-level fire detection on Wildfire-DB (ROC-AUC 0.8200, PR-AUC 0.8044, and transferred-threshold F1 0.7596). For tracking-oriented deployment requiring oriented localization, YOLO11-OBB provides the most reliable cross-dataset behavior among OBB-capable models while remaining computationally feasible. To analyze the feasibility of UAV deployment, we further measure inference efficiency using synchronized GPU and CPU power logs on a fixed workload of 1569 frames. YOLO-family models process the video in 5.73–12.47 seconds with net energy of 1247.28–1775.39 J, substantially lower latency and energy than heavier classification and reconstruction baselines. Overall, model optimality depends on operational objectives: YOLOv6 is best for cross-dataset detection robustness, whereas YOL...

Color segmentation↗
DOE OSTI2021

A Review on Bioinspired Proton Exchange Membrane Fuel Cell: Design and Materials

With the climate crisis gathering recognition, there has been significant interest in clean energy. As a reliable and clean energy solution, markets transitioning toward sustainable energy have identified polymer electrolyte membrane fuel cells (PEMFCs) as a critical technology. The increasing interest in PEMFCs is primarily due to their high energy density, high efficiency and zero greenhouse gas emission. However, additional development is required to overcome the current barriers associated with fuel cell component design, high manufacturing cost, and insufficient stability. Bioinspired designs have the potential to provide new and valuable insights using novel techniques for extracting design inspirations from biological structures. Applying these design concepts to develop materials and structures for various fuel cell components can be transformative. The well–adapted structures and functional features of biological systems have been proven to greatly enhance the performance of current designs through novel and optimized pathways. Bioinspired designs have indeed lived up to performance expectations and even exceed them. Herein, the potential and value of using these bioinspired designs are discussed. Recent studies and advancements based on nature–inspired structures for components of PEMFCs and how these design considerations substantiate their claim as an attractive alternative to conventional designs are discussed.

25 ENERGY STORAGE↗
DOE OSTI2024

KOMPASS-II: Compaction of Crushed salt for Safe Containment – Phase 2

Long-term stable sealing elements are a basic component in the safety concept for a possible repository for heat-emitting radioactive waste in rock salt. The sealing elements will be part of the closure concept for drifts and shafts. They will be made from a welldefinied crushed salt in employ a specific manufacturing process. The use of crushed salt as geotechnical barrier as required by the German Site Selection Act from 2017 /STA 17/ represents a paradigm change in the safety function of crushed salt, since this material was formerly only considered as stabilizing backfill for the host rock. The demonstration of the long-term stability and impermeability of crushed salt is crucial for its use as a geotechnical barrier. The KOMPASS-II project, is a follow-up of the KOMPASS-I project and continues the work with focus on improving the understanding of the thermal-hydraulic-mechanical (THM) coupled processes in crushed salt compaction with the objective to enhance the scientific competence for using crushed salt for the long-term isolation of high-level nuclear waste within rock salt repositories. The project strives for an adequate characterization of the compaction process and the essential influencing parameters, as well as a robust and reliable long-term prognosis using validated constitutive models. For this purpose, experimental studies on long-term compaction tests are combined with microstructural investigations and numerical modeling. The long-term compaction tests in this project focused on the effect of mean stress, deviatoric stress and temperature on the compaction behavior of crushed salt. A laboratory benchmark was performed identifying a variability in compaction behavior. Microstructural investigations were executed with the objective to characterize the influence of pre-compaction procedure, humidity content and grain size/grain size distribution on the overall compaction process of crushed salt with respect to the deformation mechanisms. The created database was used for benchmark calculations aiming for improvement and optimization of a large number of constitutive models available for crushed salt. The models were calibrated, and the improvement process was made visible applying the virtual demonstrator.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2024

Multiphysics Co-Optimization Design and Analysis of Double-Side Cooled Silicon Carbide-Based Power Module: Preprint

With the rapid growth of Electric Vehicles (EVs) and Hybrid Electric Vehicles (HEVs), much more rigorous design targets have been set for automotive power electronics, including high power density, high reliability, and low cost. Novel power module and inverter technologies based on wide bandgap (WEG) semiconductors have been developed to meet these design targets, while providing optimal power semiconductor operating temperature and promising thermomechanical performance. Compared with conventional cooling techniques which are normally applied only on one side of power module, double-side cooling approach is now believed to be the solution to enable high power density and low thermal resistance of WEG semiconductor-based power electronics. In this work, we develop a three-phase power module that is double-sided cooled using dielectric fluid jet impingement. In each phase, four silicon carbide (SiC) power semiconductors are bonded to copper busbars without electrical insulation layers. A finite element analysis (FEA) model is created for thermal and thermomechanical analysis. Based on FEA modeling results, we select particular dimensions for a parametric study to optimize thermal and mechanical performance. Using a multi-objective genetic algorithm (MOGA)-based optimization method, we have minimized the maximum junction temperature and thermal stresses within the power module. The multiphysics co-optimization approach has enabled an efficient design process of power modules with greatly reduced computational cost, as compared to conventional processes that rely on exhaustive numerical simulations and iterations.

ADVANCED PROPULSION SYSTEMS↗
NASA NTRS2004

Vibration-Based Data Used to Detect Cracks in Rotating Disks

Rotor health monitoring and online damage detection are increasingly gaining the interest of aircraft engine manufacturers. This is primarily due to the fact that there is a necessity for improved safety during operation as well as a need for lower maintenance costs. Applied techniques for the damage detection and health monitoring of rotors are essential for engine safety, reliability, and life prediction. Recently, the United States set the ambitious goal of reducing the fatal accident rate for commercial aviation by 80 percent within 10 years. In turn, NASA, in collaboration with the Federal Aviation Administration, other Federal agencies, universities, and the airline and aircraft industries, responded by developing the Aviation Safety Program. This program provides research and technology products needed to help the aerospace industry achieve their aviation safety goal. The Nondestructive Evaluation (NDE) Group of the Optical Instrumentation Technology Branch at the NASA Glenn Research Center is currently developing propulsion-system-specific technologies to detect damage prior to catastrophe under the propulsion health management task. Currently, the NDE group is assessing the feasibility of utilizing real-time vibration data to detect cracks in turbine disks. The data are obtained from radial blade-tip clearance and shaft-clearance measurements made using capacitive or eddy-current probes. The concept is based on the fact that disk cracks distort the strain field within the component. This, in turn, causes a small deformation in the disk's geometry as well as a possible change in the system's center of mass. The geometric change and the center of mass shift can be indirectly characterized by monitoring the amplitude and phase of the first harmonic (i.e., the 1 component) of the vibration data. Spin pit experiments and full-scale engine tests have been conducted while monitoring for crack growth with this detection methodology. Even so, published data are extremely limited, and the basic foundation of the methodology has not been fully studied. The NDE group is working on developing this foundation on the basis of theoretical modeling as well as experimental data by using the newly constructed subscale spin system shown in the preceding photograph. This, in turn, involved designing an optimal sub-scale disk that was meant to represent a full-scale turbine disk; conducting finite element analyses of undamaged and damaged disks to define the disk's deformation and the resulting shift in center of mass; and creating a rotordynamic model of the complete disk and shaft assembly to confirm operation beyond the first critical concerning the subscale experimental setup. The finite element analysis data, defining the center of mass shift due to disk damage, are shown. As an example, the change in the center of mass for a disk spinning at 8000 rpm with a 0.963-in. notch was 1.3 x 10(exp -4) in. The actual vibration response of an undamaged disk as well as the theoretical response of a cracked disk is shown. Experiments with cracked disks are continuing, and new approaches for analyzing the captured vibration data are being developed to better detect damage in a rotor. In addition, the subscale spin system is being used to test the durability and sensitivity of new NDE sensors that focus on detecting localized damage. This is designed to supplement the global response of the crack-detection methodology described here.

Gyekenyesi, Andrew L.↗
NASA NTRS2021

Orion Meteoroid and Debris Analysis with a Single Kevlar® Layer Enhancement to the Propulsion Tank Multi-Layer Insulation: Comparison of the Observed Data and the Adjusted Ballistic Models

In a continual mass optimization effort for the Orion vehicle by NASA and its prime contractor Lockheed Martin Space, along with the European Space Agency (ESA) and its contractors Airbus and Thales Alenia Space-Italy (TAS-I), a series of shield characterization shots have been performed to quantify the ballistic enhancement of a single-layer of Kevlar® in multi-layer insulation (MLI). This enhanced MLI is part of Orion’s propulsion subsystem, and it is planned to replace a double-layer Kevlar® configuration starting at the Artemis IV mission. This effort has included fifty-two shield characterization shots for five different shield configurations over the four Orion propellant storage vessels. These storage vessels are located in the European Service Module (ESM). This shield characterization has been managed by the Hypervelocity Impact Technology (HVIT) group in NASA Astromaterials Research and Exploration Sciences at Johnson Space Center (JSC) for the NASA Engineering and Safety Center (NESC) and the Multi-Purpose Crew Vehicle (MPCV) program office in coordination with Lockheed Martin Space and Thales Alenia Space-Italy (TAS-I). The shield characterization has used the two-stage, light-gas-gun at the Remote Hypervelocity Test Laboratory (RHTL) of NASA JSC White Sands Test Facility (WSTF) for acceleration of projectiles to representative orbital impact speeds. From this shield characterization effort, a revised ballistic limiting equation (BLE) has been developed for each of the five representative shielding configurations of the ESM pressure vessels. This document provides a brief description of the shield configurations considered for the Orion program and the shield characterization findings. These results are compared to the assumed BLE for each configuration going into the Artemis missions that utilize this version of ballistic enhanced MLI. The models are integrated into a fourteen day High Lunar Orbit (HLO) mission as a representative reliability prediction impact for Orion, and it has been found that the predicted risk of a loss-of-crew (LOC) event is actually a little less than half of the assumed value that launched this design modification opportunity.

Joshua E Miller↗
NASA NTRS2017

Current Safety Nets Within the U.S. National Airspace System

There are over 70,000 flights managed per day in the National Airspace System, with approximately 7,000 aircraft in the air over the United States at any given time. Operators of each of these flights would prefer to fly a user-defined “4D” trajectory (4DT), which includes arrival and departure times; preferred gates and runways at the airport; efficient, wind-optimal routes for departure, cruise and arrival phase of flight; and fuel efficient altitude profiles. To demonstrate the magnitude of this achievement a single flight from Los Angeles to Baltimore, accesses over 35 shared or “constrained” resources that are managed by roughly 30 air traffic controllers (at towers, approach control and en route sectors); along with traffic managers at 12 facilities, using over 22 different, independent automation system (including TBFM, ERAM, STARS, ASDE-X, FSM, TSD, GPWS, TCAS, etc.). In addition, dispatchers, ramp controllers and others utilize even more systems to manage each flight’s access to operator-managed resources. Flying an ideal 4DT requires successful coordination of all flight constraints among all flights, facilities, operators, pilots and controllers. Additionally, when conditions in the NAS change, the trajectories of one or more aircraft may need to be revised to avoid loss of flight efficiency, predictability, separation or system throughput. The Aviation Safety Network has released the 2016 airliner accident statistics showing a very low total of 19 fatal airliner accidents, resulting in 325 fatalities1. Despite several high profile accidents, the year 2016 turned out to be a very safe year for commercial aviation, Aviation Safety Network data show. Over the year 2016 the Aviation Safety Network recorded a total of 19 fatal airliner accidents [1], resulting in 325 fatalities. This makes 2016 the second safest year ever, both by number of fatal accidents as well as in terms of fatalities. In 2015 ASN recorded 16 accidents while in 2013 a total of 265 lives were lost. How can we keep it that way and not upset the apple cart by premature insertion of innovative technologies, functions, and procedures? In aviation, safety nets function as the last system defense against incidents and accidents. Current ground-based and airborne safety nets are well established and development to make them more efficient and reliable continues. Additionally, future air traffic control safety nets may emerge from new operational concepts.

Smith, Brian E.↗
DOE OSTI2024

EHR-BERT: A BERT-based model for effective anomaly detection in electronic health records

Objective: Physicians and clinicians rely on data contained in electronic health records (EHRs), as recorded by health information technology (HIT), to make informed decisions about their patients. The reliability of HIT systems in this regard is critical to patient safety. Consequently, better tools are needed to monitor the performance of HIT systems for potential hazards that could compromise the collected EHRs, which in turn could affect patient safety. In this paper, we propose a new framework for detecting anomalies in EHRs using sequence of clinical events. This new framework, EHR-Bidirectional Encoder Representations from Transformers (BERT), is motivated by the gaps in the existing deep-learning related methods, including high false negatives, sub-optimal accuracy, higher computational cost, and the risk of information loss. EHR-BERT is an innovative framework rooted in the BERT architecture, meticulously tailored to navigate the hurdles in the contemporary BERT method; thus, enhancing anomaly detection in EHRs for healthcare applications.Methods: The EHR-BERT framework was designed using the Sequential Masked Token Prediction (SMTP) method. This approach treats EHRs as natural language sentences and iteratively masks input tokens during both training and prediction stages. This method facilitates the learning of EHR sequence patterns in both directions for each event and identifies anomalies based on deviations from the normal execution models trained on EHR sequences.Results: Extensive experiments on large EHR datasets across various medical domains demonstrate that EHR-BERT markedly improves upon existing models. It significantly reduces the number of false positives and enhances the detection rate, thus bolstering the reliability of anomaly detection in electronic health records. This improvement is attributed to the model’s ability to minimize information loss and maximize data utilization effectively.Conclusion: EHR-BERT showcases immense potential in decreasing medical errors related to anomalous clinical events, positioning itself as an indispensable asset for enhancing patient safety and the overall standard of healthcare services. The framework effectively overcomes the drawbacks of earlier models, making it a promising solution for healthcare professionals to ensure the reliability and quality of health data.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2024

Interparticle Characterization of Mechanical Biomass Particle-Particle and Particle-Wall Interactions

The biomass materials industry faces significant challenges in managing material variability and its impact on storage and handling systems. Physical properties such as moisture content, particle size, and density fluctuate considerably, leading to operational issues like bridging and ratholing that disrupt material flow. These variations create a complex cascade effect throughout the process chain, affecting transportation, storage, and conversion processes. The economic consequences of this variability manifest in increased operational costs, maintenance requirements, and system downtime. Environmental factors further complicate the situation, as weather conditions and seasonal availability influence material properties and system performance. Engineers employ specialized equipment design, material characterization protocols, and pre-processing steps like size reduction and homogenization to address these challenges. A critical knowledge gap exists between continuous-level constitutive models and particle-scale behavior. This project developed a novel device to quantify interparticle mechanics between biomass particles, measuring friction and adhesion forces between particles and wall materials. The research focused on corn stover and southern pine forest residue, creating a comprehensive database of particle interactions. This breakthrough enables direct application in particle-based computational modeling, advancing the field's understanding of biomass handling characteristics and supporting the development of more reliable and efficient storage and handling systems. The project's outcomes contribute significantly to understanding biomass's mechanical and flow characteristics, particularly how variability at the particle level affects larger-scale handling operations. This knowledge is crucial for engineering feedstock supply systems that consistently meet quality and cost specifications for various conversion processes. The innovative experimental setup developed through this research represents a significant advancement in biomass characterization methodology. Providing precise measurements of particle-level interactions establishes a foundation for more accurate predictive modeling of bulk material behavior. This enhanced understanding of fundamental particle mechanics enables engineers to anticipate better and address handling challenges before they manifest in full-scale operations. This research opens new avenues for optimizing biomass handling systems through data-driven design approaches. The comprehensive database of particle interactions serves as a valuable resource for future research and development efforts, potentially leading to more efficient and cost-effective biomass processing solutions. This advancement in particle-level mechanics could revolutionize how biomass handling systems are designed and operated, contributing to more sustainable and reliable renewable energy production.

09 BIOMASS FUELS↗
DOE OSTI2023

A Self-Sustained CPS Design for Reliable Wildfire Monitoring

Continuous monitoring of areas nearby the electric grid is critical for preventing and early detection of devastating wildfires. Existing wildfire monitoring systems are intermittent and oblivious to local ambient risk factors, resulting in poor wildfire awareness. Ambient sensor suites deployed near the gridlines can increase the monitoring granularity and detection accuracy. However, these sensors must address two challenging and competing objectives at the same time. First, they must remain powered for years without manual maintenance due to their remote locations. Second, they must provide and transmit reliable information if and when a wildfire starts. The first objective requires aggressive energy savings and ambient energy harvesting, while the second requires continuous operation of a range of sensors. To the best of our knowledge, this paper presents the first self-sustained cyber-physical system that dynamically co-optimizes the wildfire detection accuracy and active time of sensors. The proposed approach employs reinforcement learning to train a policy that controls the sensor operations as a function of the environment (i.e., current sensor readings), harvested energy, and battery level. Here, the proposed cyber-physical system is evaluated extensively using real-life temperature, wind, and solar energy harvesting datasets and an open-source wildfire simulator. In long-term (5 years) evaluations, the proposed framework achieves 89% uptime, which is 46% higher than a carefully tuned heuristic approach. At the same time, it averages a 2-minute initial response time, which is at least 2.5× faster than the same heuristic approach. Furthermore, the policy network consumes 0.6 mJ per day on the TI CC2652R microcontroller using TensorFlow Lite for Micro, which is negligible compared to the daily sensor suite energy consumption.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

WISP: Watching grid Infrastructure Stealthily through Proxies (Final Technical Report)

The complex interdependencies of cyber systems (sensors and communications), physical grids and associated electricity market operations make protecting electric power grids a significant challenge. The energy sector is constantly under new, targeted, advanced and dangerous cyber-attacks that have the potential to result in the loss of human life. These threats are further exacerbated by our need to modernize the grid. One focus of cyber security research in smart grids is the securing of the SCADA system through advanced intrusion detection systems (IDS) and bad data detection algorithms in state estimation. These methods either require full knowledge of the system topology and parameters or fail to understand the physical behaviors under attack. WISP (Watching grid Infrastructure Stealthily through Proxies) is designed to provide additional protection to the power grid using only publicly available data. In particular, WISP exploits the spatio-temporal nature of the real time locational marginal prices (LMPs), in conjunction with other information such as bids, weather, outages and load data to analyze anomalous power pricing behaviors and then correlate those observations to localize regions of interest and identify potential cyber events. WISP is non-intrusive as the tool is deployed as a service in the Cloud or on premise and provides reliable information to system operators for enhanced situational awareness, without impeding energy delivery functions. The WISP technology comprises three modules: the data-driven anomaly detection core, the vulnerability and risk analysis and the root cause analysis. The data-driven anomaly detection core performs the tasks of feature selection, anomaly detection and attack region localization. The vulnerability and risk analysis module provides system level information of the vulnerable variables and times, assisting the operators in selecting monitoring and protection nodes. The root cause analysis module takes the detection results and identifies potential operational conditions that contribute to the detected anomalies. In Phase I, we have demonstrated the feasibility and effectiveness of WISP. We developed a realistic electricity market simulator capable of generating normal and attack market data under various operational conditions. We developed a series of cyber-attack detection and analysis algorithms and evaluated them under multiple data sources. Finally, we integrated all modules into an end-to-end software, providing functions for data management, data analytics and visualization. Specifically, we have achieved: (i) real-time data acceptance from external utility interfaces with >99% acceptance rate; (ii) high performance anomaly detection algorithms with >98% detection accuracy and <0.1% false alarm rate; and (iii) ultra-low computing delay <50 milliseconds. Additionally, our team developed algorithms to identify the vulnerable variables in electricity market operations and root cause analysis functions to identify major contributors to the price spikes. These ancillary modules are necessary when deploying WISP in real world industry environment. In Phase II, we have demonstrated the effectiveness of WISP software on realistic largescale power systems. We performed red team testing for the Phase I WISP software and identified software vulnerabilities and implemented corresponding mitigation solutions. We adapted the electricity market simulator for the Texas synthetic 2000-bus system and generated datasets for the false data injection attacks. We created database and visualization interfaces for the Texas system and the ISO New England system. We performed software optimization in terms of operation efficiency, computing speed and detection accuracy. Finally, we tested the software on the Texas system and the ISO New England system and evaluated the detection performance. Overall, we achieved above 89% detection rate, below 3% false alarm rate and below 37 seconds of end-to-end detection delay.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Thermal Optimization and Experimental Demonstration of a Silicon Carbide, Half-Bridge Power Module

This paper describes modeling and experimental work to design and demonstrate a silicon carbide (SiC), single-side-cooled power module. Two SiC metal-oxide-semiconductor field-effect transistors (MOSFET) were used for each switch position and packaged on a metalized ceramic substrate. Thermal modeling was first conducted to optimize the module packaging to minimize junction-to-fluid thermal resistance and device temperature variations. This procedure identified the optimal device placement and baseplate thickness to maximize thermal performance. A heat exchanger - incorporating impinging jets of water-ethylene glycol - was designed to cool the module's baseplate to minimize thermal resistance and pumping power. The heat exchanger used relatively low fluid velocities (<1.5 m/s) to minimize erosion-corrosion effects and relatively large fluid channel sizes (>=1 mm) to prevent clogging. The power module concept was fabricated, and experiments were conducted to measure its thermal performance. Direct current was used to power the module devices, and junction temperatures were measured via device temperature-sensitive parameters. The experimental results were found to be in good agreement with model predictions (within 2%) and yielded junction-to-fluid thermal resistance values as low as about 15 mm2 K/W at relatively low pumping power values. The results demonstrate the improved performance associated with jet impingement cooling and represent an improvement over existing channel-flow-type cold plates, which can enable improvements to power density and reliability.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Comprehensive assessment of deep reinforcement learning approaches for economic dispatch in nuclear-driven microgrids

As the electrical grid integrates more variable renewable energy sources such as wind and solar, the demand for distributed and flexible systems to address this increased variability becomes critical. Nuclear-driven microgrids provide a promising solution by offering stable generation to complement intermittent renewables, ensuring grid reliability and operating efficiency. This paper proposes a recurrent deep reinforcement learning framework for optimal economic dispatch in a nuclear-powered microgrid integrating renewable energy sources, small modular reactors, battery storage systems, and balance-of-plant dynamics. A three-agent control architecture is developed, where demand and renewable energy agents act as forecasters, and a reinforcement learning-based dispatch agent performs real-time energy allocation. A nonlinear programming formulation is first used to generate an optimal baseline for benchmarking. The proposed dispatch controller, based on Proximal Policy Optimization enhanced with Long Short-Term Memory networks, exploits temporal correlations in system dynamics by taking advantage of the time series used as inputs to improve policy robustness under uncertainty. Comparative analysis against established deep reinforcement learning methods, including Proximal Policy Optimization with a feedforward architecture, Soft Actor-Critic, and Twin Delayed Deep Deterministic Policy Gradient, demonstrates superior performance. Numerical results indicate that the proposed controller achieves a 0.39% cost reduction relative to the nonlinear programming benchmark and outperforms other learning-based methods by generating additional revenue of up to 0.35%. All reinforcement learning controllers compute dispatch actions in less than 0.3 s, resulting in a computational speedup of more than three orders of magnitude over the nonlinear programming baseline. The findings of this paper highlight their applicability for real-time operation and control in nuclear-integrated microgrids under volatile operating conditions.

24 POWER TRANSMISSION AND DISTRIBUTION↗