Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data analytics”

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 1,207 records · Page 67

A Digital Twin Framework Utilizing Machine Learning for Robust Predictive Maintenance: Enhancing Tire Health Monitoring

We introduce a novel digital twin (DT) framework for the predictive maintenance of long-term physical systems. Using monitoring tire health as an application, we show how the DT framework can be used to enhance automotive safety and efficiency, and how the technical challenges can be overcome using a three-step approach. First, to manage the data complexity over a long operation span, we employ data reduction techniques to concisely represent physical tires using historical performance and usage data. Relying on these data, for fast real-time prediction, we train a transformer-based model offline on our concise dataset to predict future tire health over time, represented as remaining casing potential (RCP). Based on our architecture, our model quantifies both epistemic and aleatoric uncertainties, providing reliable confidence intervals around predicted RCP. Second, to incorporate real-time data, we update the predictive model in the DT framework, ensuring its accuracy throughout its lifespan with the aid of hybrid modeling and the use of the discrepancy function. Third, to assist decision-making in predictive maintenance, we implement a tire state decision algorithm, which strategically determines the optimal timing for tire replacement based on RCP forecasted by our transformer model. This approach ensures that our DT accurately predicts system health, continually refines its digital representation, and supports predictive maintenance decisions. Furthermore, our framework effectively embodies a physical system, leveraging big data and machine learning (ML) for predictive maintenance, model updates, and decision-making.

advanced computing infrastructure↗

SU-E-I-10: Investigation On Detectability of a Small Target for Different Slice Direction of a Volumetric Cone Beam CT Image

Purpose: To investigate the detectability of a small target for different slice direction of a volumetric cone beam CT image and its impact on dose reduction. Methods: Analytic projection data of a sphere object (1 mm diameter, 0.2/cm attenuation coefficient) were generated and reconstructed by FDK algorithm. In this work, we compared the detectability of the small target from four different backprojection Methods: hanning weighted ramp filter with linear interpolation (RECON 1), hanning weighted ramp filter with Fourier interpolation (RECON2), ramp filter with linear interpolation (RECON 3), and ramp filter with Fourier interpolation (RECON4), respectively. For noise simulation, 200 photons per measurement were used, and the noise only data were reconstructed using FDK algorithm. For each reconstructed volume, axial and coronal slice were extracted and detection-SNR was calculated using channelized Hotelling observer (CHO) with dense difference-of-Gaussian (D-DOG) channels. Results: Detection-SNR of coronal images varies for different backprojection methods, while axial images have a similar detection-SNR. Detection-SNR{sup 2} ratios of coronal and axial images in RECON1 and RECON2 are 1.33 and 1.15, implying that the coronal image has a better detectability than axial image. In other words, using coronal slices for the small target detection can reduce the patient dose about 33% and 15% compared to using axial slices in RECON 1 and RECON 2. Conclusion: In this work, we investigated slice direction dependent detectability of a volumetric cone beam CT image. RECON 1 and RECON 2 produced the highest detection-SNR, with better detectability in coronal slices. These results indicate that it is more beneficial to use coronal slice to improve detectability of a small target in a volumetric cone beam CT image. This research was supported by the MSIP (Ministry of Science, ICT and Future Planning), Korea, under the IT Consilience Creative Program (NIPA-2014-H0201-14-1002) supervised by the NIPA (National IT Industry Promotion Agency). Authors declares that s/he has no conflict of Interest in relation to the work in this abstract.

60 APPLIED LIFE SCIENCES↗

Research and Education in Computational Science and Engineering

Over the past two decades the field of computational science and engineering (CSE) has penetrated both basic and applied research in academia, industry, and laboratories to advance discovery, optimize systems, support decision-makers, and educate the scientific and engineering workforce. Informed by centuries of theory and experiment, CSE performs computational experiments to answer questions that neither theory nor experiment alone is equipped to answer. CSE provides scientists and engineers of all persuasions with algorithmic inventions and software systems that transcend disciplines and scales. Carried on a wave of digital technology, CSE brings the power of parallelism to bear on troves of data. Mathematics-based advanced computing has become a prevalent means of discovery and innovation in essentially all areas of science, engineering, technology, and society; and the CSE community is at the core of this transformation. However, a combination of disruptive developments— including the architectural complexity of extreme-scale computing, the data revolution that engulfs the planet, and the specialization required to follow the applications to new frontiers—is redefining the scope and reach of the CSE endeavor. This report describes the rapid expansion of CSE and the challenges to sustaining its bold advances. The report also presents strategies and directions for CSE research and education for the next decade.

computational science and engineering↗

Bioinformatics resource manager v2.3: an integrated software environment for systems biology with microRNA and cross-species analysis tools

Background: MicroRNAs (miRNAs) are noncoding RNAs that direct post-transcriptional regulation of protein coding genes. Recent studies have shown miRNAs are important for controlling many biological processes, including nervous system development, and are highly conserved across species. Given their importance, computational tools are necessary for analysis, interpretation and integration of high-throughput (HTP) miRNA data in an increasing number of model species. The Bioinformatics Resource Manager (BRM) v2.3 is a software environment for data management, mining, integration and functional annotation of HTP biological data. In this study, we report recent updates to BRM for miRNA data analysis and cross-species comparisons across datasets. Results: BRM v2.3 has the capability to query predicted miRNA targets from multiple databases, retrieve potential regulatory miRNAs for known genes, integrate experimentally derived miRNA and mRNA datasets, perform ortholog mapping across species, and retrieve annotation and cross-reference identifiers for an expanded number of species. Here we use BRM to show that developmental exposure of zebrafish to 30 uM nicotine from 6–48 hours post fertilization (hpf) results in behavioral hyperactivity in larval zebrafish and alteration of putative miRNA gene targets in whole embryos at developmental stages that encompass early neurogenesis. We show typical workflows for using BRM to integrate experimental zebrafish miRNA and mRNA microarray datasets with example retrievals for zebrafish, including pathway annotation and mapping to human ortholog. Functional analysis of differentially regulated (p<0.05) gene targets in BRM indicates that nicotine exposure disrupts genes involved in neurogenesis, possibly through misregulation of nicotine-sensitive miRNAs. Conclusions: BRM provides the ability to mine complex data for identification of candidate miRNAs or pathways that drive phenotypic outcome and, therefore, is a useful hypothesis generation tool for systems biology. The miRNA workflow in BRM allows for efficient processing of multiple miRNA and mRNA datasets in a single software environment with the added capability to interact with public data sources and visual analytic tools for HTP data analysis at a systems level. BRM is developed using Java™ and other open-source technologies for free distribution (http://www.sysbio.org/dataresources/brm.stm).

59 BASIC BIOLOGICAL SCIENCES↗

Use of a New Analytical Model To Match Production Data and Identify Opportunities To Maximize Well Productivity in the Tuscaloosa Marine Shale Reservoir

The low productivity of the oil wells in the Tuscaloosa Marine Shale (TMS) Trend, located in Louisiana and Mississippi, is a mystery. Production data from 55 wells in the TMS Trend were analyzed to identify possible means of enhancing well productivity. The -1/2 slope in a log–log plot for reservoir linear flow (RLF) was observed for some wells, but not all, and the -1 slope for boundary–dominated flow (BDF) has not been seen in the past 5 years of production. The behavior of the TMS wells is attributed to the ultralow permeability of the TMS matrix and the decline of fracture conductivity, both of which delay the BDF. On the basis of the concept of the distance of investigation, derived from the radius–of–investigation concept, the matrix permeability of the seven TMS wells was estimated to be between 34 and 65 nd, with an average of 51 nd. A new mathematical model for production decline of multifractured horizontal wells was developed taking into consideration the time–dependent fracture conductivity during the BDF period. This model fits TMS–well production data, with an average error of 6.12% and an R2 value of 0.96. Assuming constant matrix permeability, fitting the new model to the production data in the BDF period gives a rate of fracture–conductivity decline of between 0.20 and 0.74% per month. The new mathematical model reveals opportunities to optimize well completion to enhance well productivity in the TMS Trend. For a given amount of fracture proppant allocated to a well, the shortest possible fracture spacing should be used to maximize well productivity. If the number of fractures is fixed as a constraint of well completion design, well productivity is inversely proportional to the square root of fracture spacing (i.e., if the fracture spacing is shortened by fourfold, well productivity is expected to double). If the horizontal wellbore length is fixed as a constraint of well completion, well productivity is inversely proportional to fracture spacing (i.e., if the fracture spacing is shortened by 50%, well productivity is expected to double).

Engineering↗

Corrective Action Decision Document/Closure Report for Corrective Action Unit 367: Area 10 Sedan, Ess and Uncle Unit Craters Nevada National Security Site, Nevada, Revision 0 with ROTC 1

Corrective Action Unit 367 comprises four corrective action sites (CASs): • 10-09-03, Mud Pit • 10-45-01, U-10h Crater (Sedan) • 10-45-02, Ess Crater Site • 10-45-03, Uncle Crater Site The purpose of this Corrective Action Decision Document/Closure Report is to provide justification and documentation of the corrective actions and site closure activities implemented at CAU 367. A corrective action of closure in place with use restrictions was completed at each of the three crater CASs (10-45-01, 10-45-02, and 10-45-03); corrective actions were not required at CAS 10-09-03. In addition, a limited soil removal corrective action was conducted at the location of a potential source material release. Based on completion of these correction actions, no additional corrective action is required at CAU 367, and site closure is considered complete. Corrective action investigation (CAI) activities were performed from February 2010 through March 2011, as set forth in the Corrective Action Investigation Plan for Corrective Action Unit 367: Area 10 Sedan, Ess and Uncle Unit Craters, Nevada Test Site, Nevada. The approach for the CAI was divided into two facets: investigation of the primary release of radionuclides, and investigation of non-test or other releases (e.g., migration in washes and potential source material). Based on the proximity of the Uncle, Ess, and Sedan craters, the impact of the Sedan test on the fallout deposited from the two earlier tests, and aerial radiological surveys, the CAU 367 investigation was designed to study the releases from the three crater CASs as one combined release (primary release). Corrective Action Site 10-09-03, Mud Pit, consists of two mud pits identified at CAU 367. The mud pits are considered non-test releases or other releases and were investigated independent of the three crater CASs. The purpose of the CAI was to fulfill data needs as defined during the data quality objective (DQO) process. The CAU 367 dataset of investigation results was evaluated based on a data quality assessment. This assessment demonstrated the dataset is complete and acceptable for use in fulfilling the DQO data needs. Analytes detected during the CAI were evaluated against final action levels (FALs) established in this document. For the primary release, radiological doses exceeding the FAL of 25 millirem per year were not found to be present in the surface or shallow subsurface soil outside the default contamination boundary. However, it was assumed that radionuclides are present in subsurface media within each of the three craters (Sedan, Ess, and Uncle) due to prompt injection of radionuclides from the tests. Based on the assumption of radiological dose exceeding the FAL, corrective actions were undertaken that consisted of implementing a use restriction and posting warning signs at each crater CAS. These use restrictions were recorded in the FFACO database; the U.S. Department of Energy, National Nuclear Security Administration Nevada Site Office (NNSA/NSO) Facility Information Management System; and the NNSA/NSO CAU/CAS files. With regard to other releases, no contaminants of concern were identified at the mud pits or any of the other release locations, with one exception. Potential source material in the form of lead was found at one location. A corrective action of clean closure was implemented at this location, and verification samples indicated that no further action is necessary. Therefore, NNSA/NSO provides the following recommendations: • A Notice of Completion to NNSA/NSO is requested from the Nevada Division of Environmental Protection for closure of CAU 367. • Corrective Action Unit 367 should be promoted from Appendix III to Appendix IV of the FFACO.

54 ENVIRONMENTAL SCIENCES↗

Corrective Action Decision Document/Closure Report for Corrective Action Unit 374: Area 20 Schooner Unit Crater, Nevada National Security Site, Nevada with ROTC 1 and 2, Revision 0 with ROTCs 1,2, and 3

Corrective Action Unit 374 comprises five corrective action sites (CASs): • 18-22-05, Drum • 18-22-06, Drums (20) • 18-22-08, Drum • 18-23-01, Danny Boy Contamination Area • 20-45-03, U-20u Crater (Schooner) The purpose of this Corrective Action Decision Document/Closure Report is to provide justification and documentation supporting the recommendation that no further corrective action is needed for CAU 374 based on the implementation of corrective actions. The corrective action of closure in place with administrative controls was implemented at CASs 18-23-01 and 20-45-03, and a corrective action of removing potential source material (PSM) was conducted at CAS 20-45-03. The other CASs require no further action; however, best management practices of removing PSM and drums at CAS 18-22-06, and removing drums at CAS 18-22-08 were performed. Corrective action investigation (CAI) activities were performed from May 4 through October 6, 2010, as set forth in the Corrective Action Investigation Plan for Corrective Action Unit 374: Area 20 Schooner Unit Crater, Nevada Test Site, Nevada. The approach for the CAI was divided into two facets: investigating the primary release of radionuclides and investigating other releases (migration in washes and chemical releases). The purpose of the CAI was to fulfill data needs as defined during the data quality objective (DQO) process. The CAU 374 dataset of investigation results was evaluated based on the data quality indicator parameters. This evaluation demonstrated the dataset is acceptable for use in fulfilling the DQO data needs. Analytes detected during the CAI were evaluated against final action levels (FALs) established in this document. Radiological doses exceeding the FAL of 25 millirem per year were found to be present in the surface soil that was sampled. It is assumed that radionuclide levels present in subsurface media within the craters and ejecta fields (default contamination boundaries) at the Danny Boy and Schooner sites exceed the FAL. It is also assumed that PSM in the form of lead-acid batteries at Schooner exceeds the FAL. Therefore, corrective actions were undertaken that consist of removing PSM, where present, and implementing a use restriction and posting warning signs at the Danny Boy and Schooner sites. These use restrictions were recorded in the FFACO database; the U.S. Department of Energy, National Nuclear Security Administration Nevada Site Office (NNSA/NSO) Facility Information Management System; and the NNSA/NSO CAU/CAS files. Therefore, NNSA/NSO provides the following recommendations: • No further corrective actions are necessary for CAU 374. • A Notice of Completion to NNSA/NSO is requested from the Nevada Division of Environmental Protection for closure of CAU 374. • Corrective Action Unit 374 should be moved from Appendix III to Appendix IV of the FFACO.

54 ENVIRONMENTAL SCIENCES↗

Addendum to the 2015 Eastern Interconnect Baselining and Analysis Report

This report serves as an addendum to the report 2015 Eastern Interconnect Baselining and Analysis Report (Amidan, Follum, and Freeman, 2015). This addendum report investigates the following: the impact of shorter record lengths and of adding a daily regularization term to the date/time models for angle pair measurements, additional development of a method to monitor the trend in phase angle pairs, the effect of changing the length of time to determine a baseline, when calculating atypical events, and a comparison between quantitatively discovered atypical events and actual events.

97 MATHEMATICS AND COMPUTING↗

Technology Implementation Plan: Irradiation Testing and Qualification for Nuclear Thermal Propulsion Fuel

This document is a notional technology implementation plan (TIP) for the development, testing, and qualification of a prototypic fuel element to support design and construction of a nuclear thermal propulsion (NTP) engine, specifically its pre-flight ground test. This TIP outlines a generic methodology for the progression from non-nuclear out-of-pile (OOP) testing through nuclear in-pile (IP) testing, at operational temperatures, flows, and specific powers, of an NTP fuel element in an existing test reactor. Subsequent post-irradiation examination (PIE) will occur in existing radiological facilities. Further, the methodology is intended to be nonspecific with respect to fuel types and irradiation or examination facilities. The goals of OOP and IP testing are to provide confidence in the operational performance of fuel system concepts and provide data to program leadership for system optimization and fuel down-selection. The test methodology, parameters, collected data, and analytical results from OOP, IP, and PIE will be documented for reference by the NTP operator and the appropriate regulatory and oversight authorities. Final full-scale integrated testing would be performed separately by the reactor operator as part of the preflight ground test.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

HOOVER

HOOVER is a distributed framework for solving streaming/dynamic graph problems where the structure of a graph changes over time based on the state of its vertices.

97 MATHEMATICS AND COMPUTING↗

FRMAC Laboratory Analysis RadResponder Gap Analysis and Web Portal Enhancements

The Federal Radiological Monitoring and Assessment Center (FRMAC) relies on accurate and defensible analytical laboratory data to support its mission. FRMAC Laboratory Analysis personnel are responsible for (1) receiving samples, (2) managing samples, and (3) providing data quality assurance. Currently, the RadResponder software application does not meet all these needs. With some modifications, RadResponder could meet the needs for sample receiving functions, but it does not meet the needs of sample management and data quality assurance functions. The FRMAC Laboratory Analysis team has discussed and reviewed the following options moving forward: Option 1: Make minor revisions to RadResponder to improve sample receiving capability, purchase and configure a commercial laboratory information management system (LIMS) to perform sample management and data quality assurance, and build an interface between RadResponder and the commercial-off-the-shelf LIMS. Option 2: Make major revisions to RadResponder for all FRMAC Laboratory Analysis functions to support required sample management and data quality assurance activities. Option 3: Create a custom-built LIMS system to interface with RadResponder. Note: All three options will require the development of a Laboratory Analysis web portal and will require funding for ongoing maintenance and training. The FRMAC Laboratory Analysis team highly recommends Option 1 as the best and most efficient path forward. Commercial-off-the-shelf LIMS products have been proven successful in the laboratory community for decades. Option 1 leverages these proven technologies and takes advantage of RadResponder's current strengths.

61 RADIATION PROTECTION AND DOSIMETRY↗

Cold Testing for Characterization of Oxide Feed Impurities in Support of the ARIES Program

The Advanced Recovery and Integrated Extraction System (ARIES) Program converts surplus plutonium metal components into an oxide powder form and packages the material for final disposition. In support of the ARIES program, over a metric ton (MT) of plutonium has been removed from surplus nuclear weapons since 1998 and packaged for long-term storage in 3013 cans. The oxide production rate is anticipated to increase substantially over the next decade in support of dilute and dispose goals. Analytical chemistry data from destructive analysis of the resulting 70+ blend lots to date has been reported for contaminants in the product oxide. Changes in the Pu product specification for Pu oxide product, induced by a different pit mix, the LANL swap program and processing Alternate (non-Pit Pu) Feed Stock (AFS) metals, will affect the level and the mix of these contaminants in the upcoming years. With these changes in feed material, it will be important to provide fast and efficient measurement of elemental impurities in the oxide to be certain that the process is under control and to monitor the anticipated changes from the historical blend lot data. Non-destructive methods need to be evaluated for the characterization of the plutonium oxide product and the quantification of the impurities at the current and anticipated levels. A feasibility study has begun to investigate the applicability of new spectroscopic methods for oxide impurity characterization. To this end, here we describe the preparation of surrogate materials to support these investigations, and the results of a feasibility study investigating the use of LIBS (Laser Induced Breakdown Spectroscopy) to characterize the surrogate materials as calibration standards.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Analytics-at-scale of Sensor Data for Digital Monitoring in Nuclear Plants (3 rd Annual Report)

Nuclear power plants collect and store large volumes of heterogeneous data from various components and systems. With recent advances in machine learning (ML) techniques, these data can be leveraged to develop diagnostic and short-term forecasting models to better predict future equipment condition. Maintenance operations can then be planned in advance whenever degraded performance is predicted, thus resulting in fewer unplanned outages and the optimization of maintenance activities. This enables lower maintenance costs and improves the overall economics of nuclear power. This report primarily focuses on developing a short-term forecasting process that leverages a feature selection process to distill large volumes of heterogeneous data and predict specific equipment parameters. A variety of feature selection methods, including Shapley Additive Explanations (SHAP) and variance inflation factor (VIF), were used to select the optimal features as inputs for three ML methods: long short-term memory (LSTM) networks, support vector regression (SVR), and random forest (RF). Each combination of model and input features was used to predict a pump bearing temperature both 1 and 24 hours in advance, based on actual plant system data. The optimal inputs for the LSTM and SVR were selected using the SHAP values, while the optimal input for the RF consisted solely of the response variable itself. Each model produced similar 1-hour-ahead predictions, with root mean square errors (RMSEs) of roughly 0.006. For the 24-hour-ahead predictions, differences could be seen between LSTM, SVR, and RF, as reflected by model performances of 0.036 ± 0.014, 0.0026 ± 0, and 0.063 ± 0.004 RMSE, respectively. As big data and continuous online monitoring become more widely available, the proposed feature selection process can be used for many applications beyond the prediction of process parameters within nuclear infrastructure. This report summarizes the Fiscal Year 2021 research progress encompassing the (1) data cleaning and feature selection necessary for ML applications; (2) development of short-term forecasting models to predict future plant process parameters for both single and multiple time steps ahead; and (3) validation of the feature selection methods and short-term forecasting models given new data from different systems.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

FY2018 Discovery Through Situational Awareness: Anomalies, Oscillations, and Classification

This technical report is a summary of work performed under the DOE GMLC program under the task titled “Discovery through Situational Awareness”. The focus of this work was to apply big data, statistical techniques to PMU data to do the following: 1) baseline PMU data and find anomalies with a focus on phase angle pairs, 2) identify the frequencies of dominant modal oscillations between different parts of a power system, and 3) identify and classify frequency events in PMU data. This work was performed on PMU data from the Western Interconnect. Many of these algorithms will be applied to wide area phase angle pair data on the Eastern Interconnect in a tool called ESAMS (Eastern Interconnect Situational Awareness Monitoring System). ESAMS was built by EPG (Electric Power Group) and PNNL and is currently being installed on the Eastern Interconnect. The results discussed in this report were made from 14 PMUs from data ranging from Oct 2016 to May 2018. The PMU data were measured 60 times each second. Fourteen phase angle pairs were included, with reactive and active power being calculated at 14 sites. All results have been de-identified for this report. The methodologies applied in the ESAMS tool are described in Section 2. A baseline of PMU data is created using the last 120 days of data. This baseline describes typical behavior as understood from the data. The baseline is then applied to the next day’s data and anomalies from that baseline are identified. Plots are created to help describe what is anomalous during that specific moment in time. These algorithms are currently being installed on the Eastern Interconnect. FY19 work will include discussing the results and findings from the ESAMS tool. A new method of identifying the frequencies of dominant modal oscillations between different parts of a power system is described in Section 3. A list of 199 frequency events provided by Bonneville Power Administration were used to create algorithms to detect frequency events and to create classification rules that can be used to identify future frequency events. Section 4 discusses the results the performance of 7 different event detection, clustering-based methods. Accuracy percentages ranged from 93.59% to 99.16%, with most methods being very fast. GBM (Gradient Boosting Machine) performed well and quickly. These 7 different methods were also tested to determine how well each built classification rules on training data and then applied those rules to identify future frequency events. Accuracy percentages ranged from 89.09% to 99.9%, with all methods performing quickly. GLM (General Linear Model) was viewed as the top performer when considering accuracy and processing time. Because the list of events consisted of 80% of them as active power events and only 20% as faults, all methods struggled to identify events as faults. Small faults were especially difficult to identify. A separate algorithm was created to help with this. Additional work will be done in FY19 on event detection using more data and more events and refinements to the methods.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Automated System-wide Event Detection and Classification Using Machine Learning on Synchrophasor Data

As the number of phasor measurement units (PMUs) deployed in a power system increases, and their data volume streamed to the control canter intensifies, operators are facing challenges related to the analysis of such data, which need to be observed and responded to as the measurements are displayed in the Control Room. Humans are generally unable to process such large amount of data efficiently and rapidly. There is an apparent need for automated ways to analyze the data, extract actionable information about occurrence of specific events, and characterize the events quickly and cost effectively. This paper discusses the use of machine learning (ML) to facilitate such tasks by providing automated, highly computationally efficient, and cost-effective ways of extracting actionable information from synchrophasor big data in real-time. We developed Big Data Smart (BDSmart) ML-based prototype tool for the Control Room use that automatically analyses data properties from synchrophasor system measurements taken across the three grid Interconnections in the USA (Western, Eastern and ERCOT). The data collected from several hundreds of PMUs located across the Interconnections over a period of two years have been made available for our extensive study. As a result, we were able to identify a number of big data properties that influence how ML methodology is applied to select, develop, train and test the data models that can eventually be used for the tool implementation. The resulting set of candidate algorithms spans unsupervised, supervised, semi-supervised and transfer-learning approaches. Many ML techniques, such as decision trees, multinomial logistic regression, feed-forward neural networks, K-nearest neighbor, multiclass support vector machine, and single and multi-channel convolutional neural networks, are implemented, and their performance is examined. We offer the results from testing the data models. The novelty of our study is in the approaches for bad data detection and mitigation, selection of a simplified feature for event detection, and data label improvements. As a result, we came up with a list of recommendations for the utilities on how to improve the PMU recording practices to cater to the future ML applications aimed at automating the analysis of synchrophasor data.

Synchrophasors, Machine Learning, System-wide Even↗

Energy Resilient Recovery in Puerto Rico: Hurricane Maria Recovery Support

This report covers microgrids as a special application of distributed energy resources (DERs) in Puerto Rico. This is one of a series of reports describing the U.S. Department of Energy (DOE) multi-laboratory efforts undertaken. To ensure the sustainable, long-term recovery of Puerto Rico's electric power grid from hurricanes Maria and Irma and to build capacity to manage future potential natural disasters in the most secure and resilient way, DOE convened experts from multiple national laboratories to develop a comprehensive set of data, models, analytic tools, and studies, considering inputs from a wide variety of stakeholder groups, to support technically sound recommendations for Puerto Rico's energy investment decisions. In Phase 1 of the multi-laboratory effort to support Puerto Rico's recovery, the National Renewable Energy Laboratory (NREL) provided the utility company Puerto Rico Electric Power Authority (PREPA) recommendations for a new framework of interconnection standards to accelerate the integration of utility-scale, transmission-connected, renewable electrical generation and energy storage that ensure cross-technology compatibility and enable high deployment levels without compromising grid reliability, safety, or security (https://www.nrel.gov/docs/fy19osti/73848.pdf). In Phase 2, NREL published a report focused on the interconnection of distributed energy resources (DERs) to the electric distribution system in Puerto Rico (https://www.nrel.gov/docs/fy21osti/77127.pdf). This report familiarized the reader with Puerto Rico's distribution infrastructure and operational practices and procedures that are relevant to DER interconnection. The report also provided considerations for streamlining the interconnection process given the expected increase in deployments resulting from Puerto Rico's renewable portfolio standard goal of 100% renewables by 2050. Accordingly, the report identifies considerations and concerns associated with the increase in intermittent generation, strategies for DER interconnection best practices, and the potential use of the latest technological solutions identified in the latest revision of the Institute of Electrical and Electronics Engineers 1547-2018 interconnection standard. Additionally, the report identified ways to improve the physical resiliency of installed DERs. This task describes how distributed energy systems may participate in and contribute to improving the resiliency of Puerto Rico's energy infrastructure in the future. This report presents an analysis of the Puerto Rico grid to illustrate the possible different microgrid and minigrid scenarios. The capability to form sustained microgrids and minigrids has the potential to provide a high level of resiliency during disturbances on the grid. DERs, on the other hand, have the capability to aid in the microgrid operation by providing grid support as well as grid-forming functionality. Other tasks in Phase 3 include voltage regulation in distribution networks using DERs and transmission-and-distribution co-simulation.

24 POWER TRANSMISSION AND DISTRIBUTION↗

HPC Analytics of Fused Thermal Plants Data to Optimize Operating Envelope

In this project, ORNL extensively reviewed the ORAP RAM data, and it guided us to develop machine learning models that can predict time to next failures and forecast failure trends, which will be useful for optimizing power plant operation strategies. More specifically, we trained multiple random forest models and evaluated the model accuracy to validate with 10+ years of historical data. In addition, we implemented a web-based graphical user interface system for the models to show how our models can be used in more intuitive ways. This proof of concept allowed exploration of model use with power plant operators in mind. Developed machine learning models will be helpful for managing risks, planning maintenance and operation, ultimately reducing the down time and increasing the service hours. For future work, there are several interesting research topics including but not limited to model enhancement, creating synergy with traditional failure modeling approaches, and data-driven actionable recommendation and suggestions.

20 FOSSIL-FUELED POWER PLANTS↗

Criticality Safety and Fuel Performance Considerations for Enrichment Above 5 Weight Percent in the Uranium Dioxide Fuel Cycle

The U.S. Nuclear Regulatory Commission (NRC) is preparing for anticipated licensing applications requesting fuel enrichments of UO 2 ceramic fuels enriched in excess of 5 weight percent (w/o) 235 U for commercial light-water reactors (LWRs). PNNL has been tasked with providing technical assistance to the NRC related to the review and approval of requests for enrichment increases above the current 5 w/o 235 U enrichment limit. This report will provide the agency with technical assistance to enhance the staff’s knowledge base and to identify developmental needs required to support the licensing of greater than 5 w/o UO 2 ceramic fuels. This report provides a background on increasing enrichment of UO 2 LWR fuel beyond 5 w/o 235 U. The primary concern of increasing enrichment is the impacts to criticality safety. This report will focus on criticality safety as it relates to enrichment, conversion of uranium hexafluoride to UO 2 , fuel assembly fabrication, transportation of fresh/spent fuel assemblies, reactor operations, and storage of fresh/spent fuel assemblies. The report provides a general summary of each of the steps in the UO 2 fuel cycle starting from transportation of enriched UF 6 through dry cask storage of spent fuel assemblies. This report also discusses other impacts of increased enrichment during reactor operation and the surrounding fuel cycle activities. Current needs regarding both data and analytical codes to support the increase above 5 w/o 235 U are identified and discussed.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗