Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Online Learning”

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 739 records · Page 41

Adaptive Sampling of Time Series During Remote Exploration

This work deals with the challenge of online adaptive data collection in a time series. A remote sensor or explorer agent adapts its rate of data collection in order to track anomalous events while obeying constraints on time and power. This problem is challenging because the agent has limited visibility (all its datapoints lie in the past) and limited control (it can only decide when to collect its next datapoint). This problem is treated from an information-theoretic perspective, fitting a probabilistic model to collected data and optimizing the future sampling strategy to maximize information gain. The performance characteristics of stationary and nonstationary Gaussian process models are compared. Self-throttling sensors could benefit environmental sensor networks and monitoring as well as robotic exploration. Explorer agents can improve performance by adjusting their data collection rate, preserving scarce power or bandwidth resources during uninteresting times while fully covering anomalous events of interest. For example, a remote earthquake sensor could conserve power by limiting its measurements during normal conditions and increasing its cadence during rare earthquake events. A similar capability could improve sensor platforms traversing a fixed trajectory, such as an exploration rover transect or a deep space flyby. These agents can adapt observation times to improve sample coverage during moments of rapid change. An adaptive sampling approach couples sensor autonomy, instrument interpretation, and sampling. The challenge is addressed as an active learning problem, which already has extensive theoretical treatment in the statistics and machine learning literature. A statistical Gaussian process (GP) model is employed to guide sample decisions that maximize information gain. Nonsta tion - ary (e.g., time-varying) covariance relationships permit the system to represent and track local anomalies, in contrast with current GP approaches. Most common GP models are stationary, e.g., the covariance relationships are time-invariant. In such cases, information gain is independent of previously collected data, and the optimal solution can always be computed in advance. Information-optimal sampling of a stationary GP time series thus reduces to even spacing, and such models are not appropriate for tracking localized anomalies. Additionally, GP model inference can be computationally expensive.

Thompson, David R.↗

OccuTherm: Occupant Thermal Comfort Inference using Body Shape Information

Thermal comfort is a decisive factor for the well-being, productivity, and overall satisfaction of commercial building occupants. Many commercial building automation systems either use a fixed zone-wide temperature set-point for all occupants or they rely on extensive sensor deployments with frequent online interaction with occupants. This results in inadequate comfort levels or significant training effort from users, respectively. However, the increasing ubiquity of cheap, depth-based occupancy tracking systems has enabled an improvement in inferential capabilities. We propose the novel system OccuTherm to model thermal comfort of occupants. We conducted a laboratory study with 77 participants to collect data for the implementation of a thermal comfort model that derives thermal comfort using the human body shape. Based on the comparison with model baselines and ablations, we show that our approach infers thermal comfort of individuals with 60 % accuracy when body shape information is taken into account; 6 % more than state-of-the-art approaches. We make our code, mobile app, datasets, and models freely available.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Classifying Seyfert Galaxies with Deep Learning

The traditional classification for a subclass of the Seyfert galaxies is visual inspection or using a quantity defined as a flux ratio between the Balmer line and forbidden line. One algorithm of deep learning is the convolution neural network (CNN), which has shown successful classification results. We build a one-dimensional CNN model to distinguish Seyfert 1.9 spectra from Seyfert 2 galaxies. We find that our model can recognize Seyfert 1.9 and Seyfert 2 spectra with an accuracy of over 80% and pick out an additional Seyfert 1.9 sample that was missed by visual inspection. We use the new Seyfert 1.9 sample to improve the performance of our model and obtain a 91% precision of Seyfert 1.9. These results indicate that our model can pick out Seyfert 1.9 spectra among Seyfert 2 spectra. We decompose the Hα emission line of our Seyfert 1.9 galaxies by fitting two Gaussian components and derive the line width and flux. We find that the velocity distribution of the broad Hα component of the new Seyfert 1.9 sample has an extending tail toward the higher end, and the luminosity of the new Seyfert 1.9 sample is slightly weaker than the original Seyfert 1.9 sample. This result indicates that our model can pick out the sources that have a relatively weak broad Hα component. In addition, we check the distributions of the host galaxy morphology of our Seyfert 1.9 samples and find that the distribution of the host galaxy morphology is dominated by a large bulge galaxy. In the end, we present an online catalog of 1297 Seyfert 1.9 galaxies with measurements of the Hα emission line.

79 ASTRONOMY AND ASTROPHYSICS↗

EPICS for small-scale laboratories with Python soft IOCs

While the Experimental Physics and Industrial Control System (EPICS) is widely used at large laboratories for slow controls and instrumentation, the deployment of a full EPICS installation can be difficult, with a steep learning curve to new users. Taking advantage of the pythonSoftIOC module, we developed an EPICS slow controls implementation for Jefferson Lab's Hall B cryotarget written entirely in Python and based on software IOCs that communicate with instruments over Ethernet. Here, this system ran successfully, interfacing with Jefferson Lab's full EPICS network, and we offer it as an example of the capabilities of pythonSoftIOC to build lightweight, yet robust and flexible instrumentation platforms that would be easily adapted for use at a small-scale laboratory. University groups can use these examples to build complete slow controls systems, from device communication to data archiving and display, using open-source, mature EPICS tools and student-friendly Python as an alternative to expensive and proprietary systems such as LabVIEW.

Computing↗

Using Digital and Web-Based Platforms to Enhance Nuclear Power Plant Knowledge Transfer - 20203

As the workforce ages, the loss of experienced personnel can result in a loss of important knowledge. Updating Electric Power Research Institute (EPRI) research and guidance with the latest technology developments and operating experience is an important method to transfer knowledge between the retiring and incoming workforce. Studies suggest that the incoming workforce is accustomed to using digital and/or internet-based platforms to find information and may have a higher comfort with these types of platforms than they have with static, written reports. Additionally, the ease and speed of updating and delivering content through digital formats may offer an effective method for knowledge transfer and combining information from multiple reports into a single access point. EPRI has recently produced an online Decommissioning Hub available to members and is working on replicating the effort in the area of radioactive waste management called the Radwaste Web References. EPRI has extensive guidance in both disciplines. The Decommissioning Hub uses a wiki format to provide a platform that allows for both the updating of that guidance resulting from new research and new operating experiences, and combining related topical information from multiple reports. It also allows users to access the information based on key words and topical searches, and can be a tool for users to input experience to share with others. The Decommissioning Hub is accessed via a web-based portal and includes topical pages; experience summaries; a question and answer capability; and a smart search function of all EPRI decommissioning reports. Use of a wiki format allows users to input new experiences or modify existing experience summaries to add additional detail. EPRI moderates all content. The experience data base has been initially populated with information available in published EPRI reports and unpublished information currently available within EPRI. Moving forward, population of the database will continue with information available within the open literature, new information developed from current decommissioning projects, and user inputs. The web site became active in late 2017, and a major update will be issued in early 2020. Larger scale updates are planned biannually, and smaller-scale updates will occur continuously. Capturing and applying lessons learned from industry experiences is a hallmark of the global nuclear industry. This is even more important in the decommissioning technology area since very few plant staff have relevant experience and thus there is a steep learning curve. Moreover, decommissioning involves a number of complex, non-routine and specialized tasks. The web-accessible and searchable experience database provides an easy-to-use resource for capturing and disseminating practical experiences from completed and ongoing decommissioning projects. The EPRI Radwaste Web References will build upon the philosophy and objectives of the Radwaste Desk References that were published in the 1990's. The purpose of the original Radwaste Desk References was to provide nuclear power plant professionals with a how-to manual for managing radioactive waste. The Radwaste Web References will adopt this original purpose and will be a wiki type web site that provide fundamental practical and theoretical information, operating experience (OE), and information about advanced technologies and methods. As much as practicable, features that allow for interaction between users to share OE, best practices, and benchmarking will be added. Where they exist and are appropriate, the various EPRI Guidelines documents on radwaste topics will be updated and incorporated into the Radwaste Web References. The first Radwaste Web Reference will be on Low Level Waste Characterization and will build upon the EPRI Low Level Waste Characterization Guidelines (EPRI Report TR-107201, 1996) and will be available in 2020. The next Radwaste Web Reference will be on the topic of liquid radioactive waste processing and management, tentatively set for unveiling in 2021. This paper summarizes the EPRI's previous experiences with web-based platforms and future plans in this area. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Recurrent Neural Network Language Models for Open Vocabulary Event-Level Cyber Anomaly Detection

Automated analysis methods are crucial aids for monitoring and defending a network to protect the sensitive or confidential data it hosts. This work introduces a flexible, powerful, and unsupervised approach to detecting anomalous behavior in computer and network logs; one that largely eliminates domain-dependent feature engineering employed by existing methods. By treating system logs as threads of interleaved ``sentences'' (event log lines) to train online unsupervised neural network language models, our approach provides an adaptive model of normal network behavior. We compare the effectiveness of both standard and bidirectional recurrent neural network language models at detecting malicious activity within network log data. Extending these models, we introduce a tiered recurrent architecture, which provides context by modeling sequences of users' actions over time. Compared to isolation forests and Principal Components Analysis (PCA), two popular anomaly detection algorithms, we observe superior performance on the Los Alamos National Laboratory Cyber Security dataset. For log-line-level predictions, our best performing character-based model provides test set anomaly scores ranking red team events in the 98th percentile on average, and yields 0.97 area under the receiver operator characteristic curve, demonstrating the strong fine-grained anomaly detection performance of this approach on open vocabulary logging sources.

Cyber Analysis, Deep Learning, language analysis, ↗

Giovanni: A System for Rapid Access, Visualization and Analysis of Earth Science Data Online

Collecting data and understanding data structures traditionally are the first steps that a user must take, before the core investigation can begin. This is a time-consuming and challenging task, especially when science objectives require users to deal with large multi-sensor data that are usually in different formats and internal structures. The Goddard Earth Sciences Data and Information Services Center (GES DISC) has created the GES DISC Interactive Online Visualization and ANalysis Infrastructure, Giovanni, to enable Web-based visualization and analysis of satellite remotely sensed meteorological, oceanographic, and hydrologic data sets, without users having to download data. The current operational Giovanni interfaces provide the capability to process a number of important satellite measurements, such as (1) ozone and other trace gases from TOMS, OMI, HALOE, and MLS; (2) air temperature, water vapor, and geopotential height from AIRS; (3) aerosols from MODIS TerrdAqua, and GOCART model; (4) precipitation from TRMM and ground measurements; (5) chlorophyll and other ocean color products from SeaWiFS and MODIS Aqua; and (6) sea surface temperature from MODIS Aqua. Depending on the input data structure, the system provides simple statistical analysis and creates time-averaged area plot, area-averaged time series, animations, Hovmoller latitude vs. time and longitude vs. time plots, as well as vertical profiles. The inter-comparison interfaces allow a user to compare observations from different instruments, to conduct anomaly analysis, and to study basic relationships between physical parameters. Giovanni handles data with different temporal and spatial resolutions and, thus, enables both regional and global long-term climate research and short-term special events investigation, as well as data validations and assessments. Because of its simplicity of usage, Giovanni is powerful and versatile, able to assist a wide range of users, from the discipline scientists conducting preliminary research in various fields, to students in the classroom learning about weather, climate, and other natural phenomena. Giovanni can be accessed from: http://disc. esfc.nasa.gov/techlab/giovanni/index.shtml

Shen, S.↗

Astronaut Photography of the Earth: A Long-Term Dataset for Earth Systems Research, Applications, and Education

The NASA Earth observations dataset obtained by humans in orbit using handheld film and digital cameras is freely accessible to the global community through the online searchable database at https://eol.jsc.nasa.gov, and offers a useful compliment to traditional ground-commanded sensor data. The dataset includes imagery from the NASA Mercury (1961) through present-day International Space Station (ISS) programs, and currently totals over 2.6 million individual frames. Geographic coverage of the dataset includes land and oceans areas between approximately 52 degrees North and South latitudes, but is spatially and temporally discontinuous. The photographic dataset includes some significant impediments for immediate research, applied, and educational use: commercial RGB films and camera systems with overlapping bandpasses; use of different focal length lenses, unconstrained look angles, and variable spacecraft altitudes; and no native geolocation information. Such factors led to this dataset being underutilized by the community but recent advances in automated and semi-automated image geolocation, image feature classification, and web-based services are adding new value to the astronaut-acquired imagery. A coupled ground software and on-orbit hardware system for the ISS is in development for planned deployment in mid-2017; this system will capture camera pose information for each astronaut photograph to allow automated, full georegistration of the data. The ground system component of the system is currently in use to fully georeference imagery collected in response to International Disaster Charter activations, and the auto-registration procedures are being applied to the extensive historical database of imagery to add value for research and educational purposes. In parallel, machine learning techniques are being applied to automate feature identification and classification throughout the dataset, in order to build descriptive metadata that will improve search capabilities. It is expected that these value additions will increase interest and use of the dataset by the global community.

Stefanov, William L.↗

Development of Short-Term Forecasting Models Using Plant Asset Data and Feature Selection

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 paper 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.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

An On-Board Off-Board Framework for Online Replanning: Applied to UAVs in Urban Environments

Autonomous systems are being used in a multitude of areas at an increasing rate and require a high level of adaptivity and intelligence to operate safely, especially under faulty conditions. This paper introduces a novel genetic algorithm tailored for UAV trajectory replanning, with an improved execution time via search space reduction based on the operating conditions of the UAV and its remaining mission. A unique characteristic of the replanning agent is its fast-start and adaptive properties, pre-seeding candidates with partial solutions and dynamically tuning elitism, crossover, and mutation rates in correspondence to the average fitness and diversity of the population. A population restart mechanism and early stopping mechanism are evaluated as well to assess their effect on solution quality and runtime. Previous work on genetic algorithms for UAV replanning were conducted with short trajectories in a small state space. Our UAV operates in a 56,000 square meter simulated urban environment, with static obstacles and a total of 53 possible waypoints. The agent increases the safety and reliability of UAV autonomy when operating under faulty conditions and when replanning is required.

Machine Learning↗

Improving deep learning-based protein distance prediction in CASP14

Abstract Motivation Accurate prediction of residue–residue distances is important for protein structure prediction. We developed several protein distance predictors based on a deep learning distance prediction method and blindly tested them in the 14th Critical Assessment of Protein Structure Prediction (CASP14). The prediction method uses deep residual neural networks with the channel-wise attention mechanism to classify the distance between every two residues into multiple distance intervals. The input features for the deep learning method include co-evolutionary features as well as other sequence-based features derived from multiple sequence alignments (MSAs). Three alignment methods are used with multiple protein sequence/profile databases to generate MSAs for input feature generation. Based on different configurations and training strategies of the deep learning method, five MULTICOM distance predictors were created to participate in the CASP14 experiment. Results Benchmarked on 37 hard CASP14 domains, the best performing MULTICOM predictor is ranked 5th out of 30 automated CASP14 distance prediction servers in terms of precision of top L/5 long-range contact predictions [i.e. classifying distances between two residues into two categories: in contact (<8 Angstrom) and not in contact otherwise] and performs better than the best CASP13 distance prediction method. The best performing MULTICOM predictor is also ranked 6th among automated server predictors in classifying inter-residue distances into 10 distance intervals defined by CASP14 according to the precision of distance classification. The results show that the quality and depth of MSAs depend on alignment methods and sequence databases and have a significant impact on the accuracy of distance prediction. Using larger training datasets and multiple complementary features improves prediction accuracy. However, the number of effective sequences in MSAs is only a weak indicator of the quality of MSAs and the accuracy of predicted distance maps. In contrast, there is a strong correlation between the accuracy of contact/distance predictions and the average probability of the predicted contacts, which can therefore be more effectively used to estimate the confidence of distance predictions and select predicted distance maps. Availability and implementation The software package, source code and data of DeepDist2 are freely available at https://github.com/multicom-toolbox/deepdist and https://zenodo.org/record/4712084#.YIIM13VKhQM. Supplementary information Supplementary data are available at Bioinformatics online.

59 BASIC BIOLOGICAL SCIENCES↗

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 ↗

The ParFlow Sandtank: An interactive educational tool making invisible groundwater visible

Physical aquifer models are a highly effective teaching tool for hydrology education, however they come with inherent limitations that include the high cost to purchase, the static configuration of the model materials, the time required to visualize hydrogeological phenomena, and the effort to reset and clean them over time. To address these and other limitations, we have developed an interactive computer simulation of a physical aquifer model called the ParFlow Sandtank. In this gamified interface, users run the simulation using a familiar web-app like interface with sliders and buttons while learning real hydrologic concepts. Our user interface allows participants to dive into the world of hydrology, understanding assumptions about model parameters such as hydraulic conductivity, making decisions about inputs to groundwater aquifer systems such as pumping rates, visualizing outputs such as stream flow, transport, and saturation, and exploring various factors that impact real environmental systems such as climate change. The ParFlow Sandtank has already been used in a variety of educational settings with more than 9,000 users per year, and we feel this emerging educational tool can be used broadly in educational environments and can be scaled-up to provide greater accessibility for students and educators. Here we present the capabilities and workflow of the ParFlow Sandtank, two use cases, and additional tools and custom templates that have been developed to support and enhance the reach of the ParFlow Sandtank.

54 ENVIRONMENTAL SCIENCES↗

MACHINE LEARNING BASED CHEMICAL EXPLOSIVE MODE ANALYSIS (ML-CEMA)

The software consists of a Machine Learning based Chemical Explosive Mode Analysis (ML-CEMA) tool for advanced computational flame diagnostics. CEMA, originally based on eigen-analysis of the local thermochemical system, is capable of identifying reaction fronts and limit phenomena such as auto-ignition and extinction in practical combustion systems, such as internal combustion engines and gas turbine combustors. However, the original CEMA is computationally expensive for large reaction mechanisms that are typically needed to describe fuel chemistry of practical large-hydrocarbon fuels. This novel ML-CEMA tool employs a ML technique to accelerate the eigen-analysis of the basic CEMA approach by orders of magnitude, thus making it suitable for practical fuels. In ML-CEMA,zero-dimensional (0D) reactors and one-dimensional (1D) premixed flames are first used to generate a large number of data points for neural network based ML training. The trained ML model is then used to perform CEMA prediction. This ML-CEMA tool has been demonstrated in canonical 0D and 1D configurations as well as highly-transient three-dimensional spray flames exhibiting multi-mode turbulent combustion, showing promising results. ML-CEMA, as a standalone tool, can be used for computationally-efficient diagnostics of massive datasets generated from both experiments and simulations. For example, based on spatially resolved measurements of a small set of reactive scalars(such as temperature, hydroxyl radical and formaldehyde), ML-CEMA can effectively identify flame fronts and rare events. ML-CEMA also provides a robust online or offline flame feature detection tool. When used for on-the-fly simulations, ML-CEMA further enables zone-adaptive combustion modeling, in which the predicted eigenvalue is used as a robust mode indicator for judicious assignment of locally-valid combustion models. This ML-CEMA based zone-adaptive model can lead to substantial computational cost savings when used for large-scale simulations of multi-mode combustion systems. Third Party Code Web Page to Download Code Web Page Location of Third Party License

Xu, Chao↗

Panel Session 2: Hot Topics in US DOE Environmental Management

The well-attended Panel Session was chaired by Martin Schneider. The four panelists did not provide presentations. Rather the session was a discussion focusing on pressing issues facing US DOE Environmental Management (EM) sites and how the program is moving forward. The session began by momentarily reflecting on a success: the three Tank Waste Treatment Facilities will soon be coming online. These three facilities will treat as much waste in two years as was treated in the past 30 years. That is a significant milestone in progress. Also, safety enhancements at WIPP are a big success. The panelists noted that there will be challenges with the victories of the startup of the Waste Treatment Facilities. Treating tank waste is the last EM mission and will help with the relationships with the stakeholders, regulators and Congress. One of the biggest challenges will be moving back to full operation mode. Although the operation will increase productivity of reducing/managing the tank waste, it will decrease life cycle cost and lessen years of cleanup. The challenge will be having people understand that EM will be operating 24/7 and that maintenance will be key in keeping the operations. EM will be working together with the regulators to show tangible progress that will be effective and efficient. By working together, they can find opportunities/technologies for streamlining and accelerating cleanup. Recent changes to the regulatory framework between DOE-EM Los Alamos Field Office and New Mexico Environmental Department have helped cleanup in Los Alamos. Every year EM and the agencies review progress and come to agreements on the next work priorities. At Savannah River Site, the Federal Facility Agreement among EM, South Carolina Department of Health and Environmental Control, and Environmental Protection Agency (EPA) has developed a small core team that work are shared by together in resolving issues that come up. The discussion moved on to how progress can be made by sharing lessons learned among the contractors. An example of how these learned lessons can be shared is that key representatives from projects across the EM sites participate in various forums to review/contribute to the work planned. Other opportunities can be by temporary assignments of DOE staff to gain experience and by monthly EM Headquarters report that provide the lessons learned across the EM sites. The panelists discussed how skyline changes are the best examples of tangible progress. Decontamination and Demolition (D and D) of structures are the easiest tangible progress that can be seen. Whereas tank waste cleanup is the hardest to see. Disposal of waste onsite versus offsite is another challenge, and timely discussions are necessary to get to the best value to the government.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DRIPS: A framework for dimension reduction and interpolation in parameter space

Reduced-order models are often used to describe the behavior of complex systems, whose simulation with a full model is too expensive, or to extract salient features from the full model’s output. We introduce a new model-reduction framework DRIPS (dimension reduction and interpolation in parameter space) that combines the offline local model reduction with the online parameter interpolation of reduced-order bases (ROBs). The offline step of this framework relies on dynamic mode decomposition (DMD) to build a low-rank linear surrogate model, equipped with a local ROB, for quantities of interest derived from the training data generated by repeatedly solving the (nonlinear) high-fidelity model for multiple parameter points. The online step consists of the construction of a parametric reduced-order model for each target/test point in the parameter space, with the interpolation of ROBs done on a Grassman manifold and the interpolation of reduced-order operators done on a matrix manifold. The DMD component enables DRIPS to model (typically low-dimensional) quantities of interest directly, without having to access the (typically high-dimensional and possibly nonlinear) operators in a high-fidelity model that governs the dynamics of the underlying high-dimensional state variables, as required in projection-based reduced-order modeling. A series of numerical experiments suggests that DRIPS yields a model reduction, which is computationally more efficient than the commonly used projection-based proper orthogonal decomposition; it does so without requiring a prior knowledge of the governing equation for quantities of interest. Furthermore, for the nonlinear systems considered, DRIPS is more accurate than Gaussian-process interpolation (Kriging).

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Expediting DECam Multimessenger Counterpart Searches with Convolutional Neural Networks

Searches for counterparts to multimessenger events with optical imagers use difference imaging to detect new transient sources. However, even with existing artifact-detection algorithms, this process simultaneously returns several classes of false positives: false detections from poor-quality image subtractions, false detections from low signal-to-noise images, and detections of preexisting variable sources. Currently, human visual inspection to remove the false positives is a central part of multimessenger follow-up observations, but when next generation gravitational wave and neutrino detectors come online and increase the rate of multimessenger events, the visual inspection process will be prohibitively expensive. We approach this problem with two convolutional neural networks operating on the difference imaging outputs. The first network focuses on removing false detections and demonstrates an accuracy of 92% on our data set. The second network focuses on sorting all real detections by the probability of being a transient source within a host galaxy and distinguishes between various classes of images that previously required additional human inspection. We find the number of images requiring human inspection will decrease by a factor of 1.5 using our approach alone and a factor of 3.6 using our approach in combination with existing algorithms, facilitating rapid multimessenger counterpart identification by the astronomical community.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

The Potential Role of Concentrating Solar Power within the Context of DOE's 2030 Solar Cost Targets

For solar electricity generating technologies to be cost competitive at a large scale with conventionally generated electricity, cost reductions are needed for both concentrating solar power (CSP) and solar photovoltaic (PV) systems. In 2011, the U.S. Department of Energy (DOE) established solar cost targets that corresponded to reducing CSP and PV prices by approximately 75% in order to achieve a levelized cost of electricity (LCOE) of $0.06 per kilowatt-hour (kWh) for both utility-scale PV and high-capacity factor CSP-TES systems in 2020. Utility-scale PV achieved its 2020 cost target in 2017, and recent estimates for the LCOE of CSP-TES with a molten-salt power tower system are approximately $0.10/kWh for projects that are expected to come online in 2020, which represents a substantial reduction since 2010 - when the LCOE for CSP-TES was around $0.21/kWh. To continue the momentum for cost reductions in solar technologies, DOE recently established cost targets for 2030 that would make solar one of the lowest-cost sources of new electricity in the United States. For CSP-based systems, the new targets correspond to an LCOE in 2030 of $0.05/kWh for a dispatchable, high-capacity factor CSP-TES plant configuration. This aggressive target would have been unimaginable a decade ago. However, building on the previously described reduction in CSP-TES costs over the past decade, recent announcements suggest the next phase of projects will continue this downward trend through lower installation costs, attractive financing, longer-duration PPAs, and the ability to capitalize on the value that the flexibility of storage brings CSP. This presentation summarizes a recent report that evaluates the potential impacts of simultaneously achieving the 2030 cost targets for PV and CSP-TES, and it includes a detailed evaluation of the role that CSP-TES could play in realizing those impacts. The scenarios in this analysis are designed to isolate and assess the potential impacts of achieving DOE's 2030 cost targets for CSP-TES and PV, and they do not reflect the potential benefits or system impacts associated with success in other DOE research programs. The low-cost solar scenarios include DOE's 2030 solar cost targets--which are represented via a roughly 50% reduction in LCOE by 2030 (from current levels) with additional cost reductions thereafter representing technology learning and/or improvements that could result from innovation - coupled with a variety of market, technology, and demand assumptions. With these assumptions, the evolution of the contiguous U.S. electricity system is evaluated with NREL's Renewable Energy Deployment System (ReEDS) model, which was specifically designed to represent the temporal and locational value of renewable generation technologies in the U.S. power system. ReEDS relies on system-wide least-cost optimization to estimate the type and location of future generation and transmission capacity. In addition, it accounts for the locational and temporal variations in variable renewable technologies, including the need for new transmission, curtailment, dynamic capacity value, and the need to hold operating reserves to account for the uncertainty and variability of these technologies. This presentation summarizes the key findings that arise from a detailed evaluation of the impacts of achieving DOE's 2030 cost targets for CSP-TES and PV systems, noting again the inherent challenges associated with modeling future scenarios of the large, complex system electricity system in the contiguous United States within the context of various market features.

14 SOLAR ENERGY↗