Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Data science”

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 1,441 records · Page 80

Theory-Guided Machine Learning in Materials Science

Materials scientists are increasingly adopting the use of machine learning tools to discover hidden trends in data and make predictions. Applying concepts from data science without foreknowledge of their limitations and the unique qualities of materials data, however, could lead to errant conclusions. The differences that exist between various kinds of experimental and calculated data require careful choices of data processing and machine learning methods. Here, we outline potential pitfalls involved in using machine learning without robust protocols. We address some problems of overfitting to training data using decision trees as an example, rational descriptor selection in the field of perovskites, and preserving physical interpretability in the application of dimensionality reducing techniques such as principal component analysis. We show how proceeding without the guidance of domain knowledge can lead to both quantitatively and qualitatively incorrect predictive models.

36 MATERIALS SCIENCE↗

Repurposing Drilling Control Diagnostics for Subsurface Edge Detection and Boundary Advisement During Planetary Drilling

Informed decision-making during lunar drilling and sampling missions will require data monitoring tools and specialized ground data systems. Accurate and updated situational awareness, with ongoing data monitoring, is critical for timely responses by to incoming science data. Traverse plans and scheduled activities may need to be flexibly changed in order to react to unexpected data or situations. Unlike (for example) Mars missions, the relative lightspeed closeness of the Moon allows for near-real-time ground processing of incoming mission and instrument data. An Apollo-class lunar regolith drill will in a sense “travel” a meter or two vertically at a given subsurface characterization site. As the drill penetrates into lunar regolith, it is likely to encounter a range of material densities, orientations, fracture toughness, and (perhaps) ice percentages. Lunar drill telemetry can provide science teams with a valuable first look into the subsurface structure, the regolith bulk properties, and constituents at each drilled site. Real-time AI-based recognition and reaction to downhole situations has been developed for automated deeper drilling on Mars and beyond. We can leverage the same knowledge bases and pattern-matching as areal-time interpreter of the subsurface, a situational awareness tool during drilling operations. We recently (Sept. 2019) demonstrated this AI drilling monitoring and analysis capability, in control of in-situ drilling and sampling operations, mounted on a KREX-2 rover in Chile’s Atacama Desert. Terrestrial automated drilling log analyses in oil exploration have used similar machine learning techniques in classifying and identifying features in drilling logs –but these typically are designed assuming a drilling fluid influencing downhole measurements and data (permeability, resistivity). Drilling models and existing AI software designed to detect and respond to drilling faults and hard materials can be repurposed, for near-real-time (ground-based) interpretation of drilling telemetry –a potentially valuable advisory tool for strata boundaries and changes in drilling parameters. On the Moon, this approach could be used to study the structure and to some extent the composition of lunar regolith vs. borehole depth, based on recognizable variations in fracture hardness, drilling energy and penetration rates while actively drilling. Since the early 2000s, a series of increasingly-capable real-time drilling telemetry interpretation and characterization software tools have been developed. These subsurface models and software tools have monitored the real-time drilling data received, and automatically identified changes in drill behavior (e.g., encountering a harder target layer, bit inclusions, drill choking due to infall downhole, and others) correlating these with subsurface structures and features. We discuss the mappings between drill borehole parameters, faults or events detected, and modeled changes in rock layer boundaries, in examples drawn from field testing at analog sites in an Arctic impact crater, Rio Tinto, and Chile’s Atacama Desert. These demonstrate how subsurface structural boundaries led to fault detections and responses by the software.

drilling advisor↗

Toward a Data Scalable Solution for Facilitating Discovery of Science Resources

Science is increasingly motivated by the need to process larger quantities of data. It is facing severe challenges in data collection, management, and processing, so much so that the computational demands of “data scaling” are competing with, and in many fields surpassing, the traditional objective of decreasing processing time. Example domains with large datasets include astronomy, biology, genomics, climate/weather, and material sciences. This paper presents a real-world use case in which we wish to answer queries pro- vided by domain scientists in order to facilitate discovery of relevant science resources. The problem is that the metadata for these science resources is very large and is growing quickly, rapidly increasing the need for a data scaling solution. We propose a system – SGEM – designed for answering graph-based queries over large datasets on cluster architectures, and we re- port performance results for queries on the current RDESC dataset of nearly 1.4 billion triples, and on the well-known BSBM SPARQL query benchmark.

aggregation↗

Data Catalog Series for Space Science and Applications Flight Missions: Descriptions of Data Sets from Geostationary and High-Altitude Scientific Spacecraft and Investigations - Volume 2B

The main purpose of the data catalog series is to provide descriptive references to data generated by space science flight missions. The data sets described include all of the actual holdings of the Space Science Data Center (NSSDC), all data sets for which direct contact information is available, and some data collections held and serviced by foreign investigators, NASA and other U.S. government agencies. This volume contains narrative descriptions of data sets from geostationary and high altitude scientific spacecraft and investigations. The following spacecraft series are included: Mariner, Pioneer, Pioneer Venus, Venera, Viking, Voyager, and Helios. Separate indexes to the planetary and interplanetary missions are also provided.

Schofield, Norman J.↗

New technologies as decision aids for the advancement of ecological risk assessment

Moore's law states that the number of transistors that can be placed on an integrated circuit doubles every two years (Moore, 1975). This has led to a steady increase in the processing power of computers over time, and technology is now enhancing and advancing software and scientific applications, which has enabled computationally intensive methods such as machine learning, data science, modeling, and simulation. The advancement of computers and data-driven algorithms is profoundly impacting people's lives. It is changing the way we work, the way we learn, and the way we interact with the world around us. Here, this editorial will discuss how scientists can benefit from the latest technology advancements and related tools by incorporating them into the ecological risk assessment (ERA) to study ecosystems as a way to create refined assessments and accelerate the turnaround times.

54 ENVIRONMENTAL SCIENCES↗

Data synthesis and display programs for wave distribution function analysis

At the National Space Science Data Center (NSSDC) software was written to synthesize and display artificial data for use in developing the methodology of wave distribution analysis. The software comprises two separate interactive programs, one for data synthesis and the other for data display.

Storey, L. R. O.↗

Maximizing Earth Science Observations with Data Harmonization: Harmonized Landsat/Sentinel-2

In August 2021, NASA released science quality harmonized Landsat/Sentinel-2 products for both cloud-based access and direct download from the Land Processes Distributed Active Archive Center (LP DAAC). These 30-meter products, HLSS30 (Sentinel-2 component) and HLSL30 (Landsat component) are placed on the same grid and are generated and distributed fully in the cloud. Similarly, ESA is prototyping harmonized Landsat Sentinel-2 products at 10-meter resolution. The data production and science teams from NASA and ESA will present technical details on the data production system, data product status and availability, and the benefit of harmonizing instruments with similar sensing characteristics between the agencies.

Earth Science↗

Online evolutionary neural architecture search for multivariate non-stationary time series forecasting

Time series forecasting (TSF) is one of the most important tasks in data science. TSF models are usually pre-trained with historical data and then applied on future unseen datapoints. However, real-world time series data is usually non-stationary and models trained offline usually face problems from data drift. Models trained and designed in an offline fashion can not quickly adapt to changes quickly or be deployed in real-time. To address these issues, this work presents the Online NeuroEvolution-based Neural Architecture Search (ONE-NAS) algorithm, which is a novel neural architecture search method capable of automatically designing and dynamically training recurrent neural networks (RNNs) for online forecasting tasks. Without any pre-training, ONE-NAS utilizes populations of RNNs that are continuously updated with new network structures and weights in response to new multivariate input data. ONE-NAS is tested on real-world, large-scale multivariate wind turbine data as well as the univariate Dow Jones Industrial Average (DJIA) dataset. These results demonstrate that ONE-NAS outperforms traditional statistical time series forecasting methods, including online linear regression, fixed long short-term memory (LSTM) and gated recurrent unit (GRU) models trained online, as well as state-of-the-art, online ARIMA strategies. Additionally, results show that utilizing multiple populations of RNNs which are periodically repopulated provide significant performance improvements, allowing this online neural network architecture design and training to be successful.

97 MATHEMATICS AND COMPUTING↗

Global Carbon Project: the 2013 global carbon budget (includes V.1.1, Nov2013, V.1.3, Dec2013, V.2.3, June2014, and V.2.4, July2014)

Global emissions of carbon dioxide from the combustion of fossil fuels will reach 36 billion tonnes for the year 2013. "This is a level unprecedented in human history," says CSIRO's Dr Pep Canadell, Executive-Director of the Global Carbon Project (GCP) and co-author of a new report. The GCP provides an annual report of carbon dioxide emissions, land and ocean sinks and accumulation in the atmosphere, incorporating data from multiple research institutes from around the world. The 2013 figures coincide with the global launch of the Global Carbon Atlas, an online platform to explore, visualise and interpret the emissions data at the global, regional and national scales (www.globalcarbonatlas.org). The full data and methods are published today in the journal Earth System Science Data Discussions, and data and other graphic materials can be found at: www.globalcarbonproject.org/carbonbudget. The Global Carbon Budget 2013 is a collaborative effort of the global carbon cycle science community coordinated by the Global Carbon Project.

54 ENVIRONMENTAL SCIENCES↗

Global Carbon Project: the 2013 Global Carbon Budget (Version 2.3, issued June 2014)

Global emissions of carbon dioxide from the combustion of fossil fuels will reach 36 billion tonnes for the year 2013. "This is a level unprecedented in human history," says CSIRO's Dr Pep Canadell, Executive-Director of the Global Carbon Project (GCP) and co-author of a new report. The GCP provides an annual report of carbon dioxide emissions, land and ocean sinks and accumulation in the atmosphere, incorporating data from multiple research institutes from around the world. The 2013 figures coincide with the global launch of the Global Carbon Atlas, an online platform to explore, visualise and interpret the emissions data at the global, regional and national scales (www.globalcarbonatlas.org). The full data and methods are published today in the journal Earth System Science Data Discussions, and data and other graphic materials can be found at: www.globalcarbonproject.org/carbonbudget. The Global Carbon Budget 2013 is a collaborative effort of the global carbon cycle science community coordinated by the Global Carbon Project. The landing page for this dataset includes links to V. 1.1, issued Nov2013, V.1.3, issued Dec2013, and the June 2014 issue of V.2.3 of the 2013 Global Carbon Budget.

54 ENVIRONMENTAL SCIENCES↗

Challenges for monitoring and data analytics in a leadership public data repository

The availability and disposition of data has assumed increasing importance in large-scale computational science. Data repositories are evolving to meet new classes of requirements: compliance with government access guidelines, support for reproducibility of experimental results, and long-term availability of data products. The Constellation public data repository at the Oak Ridge Leadership Computing Facility faces these issues while being situated in one of the most productive data centers in the world. While monitoring and operational data analysis are ingrained in the operation of the OLCF’s large-scale high performance computing platforms, data repositories do not have this history of support. Problems faced by Constellation range from data size (over 7 petabytes in current holdings) to analytic complexity (detailed curation is both absolutely necessary for many data sets and absolutely impossible for humans to accomplish in any practical manner) to deployment environment (OLCF storage resources are oriented toward the needs of the compute platforms). In this paper we describe some of the challenges for collecting monitoring and analytic data from a leadership public data repository. We also discuss various strategies we are pursuing in order to address these challenges, from manual data collection to plans for introducing machine learning-based curatorial techniques.

Widener, Patrick [ORNL] (ORCID:0000000258820816)↗

Rational design of stealthy hyperuniform two-phase media with tunable order

Disordered stealthy hyperuniform materials are exotic amorphous states of matter that have attracted recent attention because of their novel structural characteristics (hidden order at large length scales) and physical properties, including desirable photonic and transport properties. It is therefore useful to devise algorithms that enable one to design a wide class of such amorphous configurations at will. In this paper, we present several algorithms enabling the systematic identification and generation of discrete (digitized) stealthy hyperuniform patterns with a tunable degree of order, paving the way towards the rational design of disordered materials endowed with novel thermodynamic and physical properties. To quantify the degree of order or disorder of the stealthy systems, we utilize the discrete version of the $\tau$ order metric, which accounts for the underlying spatial correlations that exist across all relevant length scales in a given digitized two-phase (or, equivalently, a two-spin state) system of interest. In this work, our results impinge on a myriad of fields, ranging from physics, materials science and engineering, visual perception, and information theory to modern data science.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A National Infrastructure for Artificial Intelligence on the Grid (NI4AI) (Final Scientific/Technical Report)

Electric utilities have traditionally taken a very pragmatic yet myopic approach with grid sensors and the resulting collected data. Sensors are purchased and deployed to solve a specific, known problem that has risen to sufficient awareness as to justify the effort of deploying sensors and the needed capital investment. This sensor data flows into proprietary software packages with limited functionality intended only to address the initial problem. This approach aligns with the financial incentives of the utility to deploy capital into fixed hardware assets for which the corporations earn a rate of return. This mentality stands in stark contrast to the big data revolution that started nearly 25 years ago with the rise of Google. In this worldview, data is a fundamental business asset; successful organizations collect, store, explore, merge, and exploit as much data as possible to not only solve problems well understood today but also to tackle new problems that will inevitably rise tomorrow. The ARPA-E Open Innovation 2018 project entitled A National Infrastructure for Artificial Intelligence on the Grid or NI4AI for short was designed to demonstrate this alternative paradigm for using data. To do this, the project was composed of three key thrust areas. The first major component deployed a variety of high-frequency grid sensors and captured terabytes of both wide-scale and localized grid measurements, generating high-value datasets for grid research and algorithm development. The second aspect made available PingThings’ PredictiveGridTM, a horizontally scalable, cloud-based data management and AI platform built for time series data to explore and exploit the collected data. Finally, the project fostered a diverse and open research community composed of experts from numerous fields through focused educational content, code sharing, and data science competitions. Shifting away from “single use” sensors and closed data silos within electric utilities is a major benefit to the public at large. This legacy approach to data is incredibly (1) capital intensive (new sensors must be deployed for each new problem and problems tend to arise continuously) and (2) painfully slow (new problems must be identified first and then new sensors must be deployed to collect data to begin to address the issue). The transition to a carbon neutral grid requires a massive transformation of the existing grid infrastructure and will continue to challenge the legacy grid in unforeseen ways. The only way to make the energy transition cost effective is for utilities to abandon this dated data paradigm and adopt more contemporary approaches. NI4AI has shown that it is technically possible and economically feasible to ingest, explore, and exploit grid data collected from even very high frequency sensing, such as continuous point on wave sensors collecting measurements 10,000 times a second. In fact, the PredictiveGrid platform used is commercially available and deployed at several utilities in the United States. Project accomplishments were numerous and included (1) making available a state of the art time series platform to the community, (2) collecting over 520 streams of time series data from grid sensors totaling over 1 trillion grid measurements, and (3) developing and nurturing a community within the industry focused on the use of data to create value for utilities and, ultimately, end consumers.

97 MATHEMATICS AND COMPUTING↗

Exploring the Whole Set of Accurate Sparse Interpretable Models

In data science applications, there are often many models that fit the data well. This phenomenon was called the Rashomon Effect by Leo Breiman. The set of good models is called the Rashomon Set, and the goal of this project is to locate, store, and study the Rashomon sets for classes of interpretable models, including decision trees and generalized additive models.

97 MATHEMATICS AND COMPUTING↗

Real-Time Lunar Prospector Data Visualization Using Web-Based Java

The Lunar Prospector was co-developed by NASA Ames Research Center and Lockheed Martin, and was launched on January 6th, 1998. Its mission is to search for water ice and various elements in the Moon's surface, map its magnetic and gravity fields, and detect volcanic activity. For the first time, the World Wide Web is being used to graphically display near-real-time data from a planetary exploration mission to the global public. Science data from the craft's instruments, as well as engineering data for the spacecraft subsystems, are continuously displayed in time-varying XY plots. The craft's current location is displayed relative to the whole Moon, and as an off-craft observer would see in the reference frame of the craft, with the lunar terrain scrolling underneath. These features are implemented as Java applets. Analyzed data (element and mass distribution) is presented as 3D lunar maps using VRML and Javascript. During the development phase, implementations of the Java Virtual Machine were just beginning to mature enough to adequately accommodate our target featureset; incomplete and varying implementations were the biggest bottleneck to our ideal of ubiquitous browser access. Bottlenecks notwithstanding, the reaction from the Internet community was overwhelmingly enthusiastic.

Deardorff, D. Glenn↗

Goddard Atmospheric Composition Data Center: Aura Data and Services in One Place

The Goddard Atmospheric Composition Data and Information Services Center (AC-DISC) is a portal to the Atmospheric Composition specific, user driven, multi-sensor, on-line, easy access archive and distribution system employing data analysis and visualization, data mining, and other user requested techniques for the better science data usage. It provides convenient access to Atmospheric Composition data and information from various remote-sensing missions, from TOMS, UARS, MODIS, and AIRS, to the most recent data from Aura OMI, MLS, HIRDLS (once these datasets are released to the public), as well as Atmospheric Composition datasets residing at other remote archive site.

Leptoukh, G.↗

Power combining considerations for project Prometheus TWTAs

NASA's Jupiter Icy Moons Orbiter (JIMO) mission would be faced with the challenging data transmission task of sending a 50 Mbps stream of science data back to Earth over a distance of 6.2 A.U. (9.3x10^8 km). This high data rate will require a transmitter power of 1 kW at 32 GHz which greatly exceeds the highest powers available from any currently existing space-qualified RF amplifiers. Initially, consideration was given to developing a space-qualified klystron or EIK because of the relatively narrow bandwidths implied by this data rate, but neither of these device types have any significant history of use in space.

power combining↗