Search NASA⌕ Search

SEARCH · Search NASA

Results for “data quality issues”

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 55 records · Page 3

Robust PCA-Deep Belief Network Surrogate Model for Distribution System Topology Identification with DERs

With the expansion of distribution networks and increased penetration of distributed energy resources (DERs), it is becoming increasingly important to obtain accurate distribution network topology in real-time. In this paper, a robust principal component analysis coupled deep belief network (PCA-DBN) surrogate model is proposed for distribution system topology identification. It integrates the benefits of robust feature extraction from PCA to deal with data quality issues and filter out noise, and the strength of DBN in capturing the nonlinear relationship between voltage amplitudes and the binary states of switchable connections. This also significantly reduces the DBN training complexity without loss of accuracy. It is shown that the widely used standard deviation of voltage drop and the voltage covariance matrix features yield less accuracy as compared to that of the voltage amplitudes in presence of high penetration of DERs and ZIP loads. Comparison results with other alternatives, such as the random forest (RF), multi-output regression (MOR) and the traditional DBN methods demonstrate that the proposed method can achieve a much higher topology identification accuracy while maintaining robustness to missing data and measurement noise under various penetration levels of DERs.

deep belief network↗

Combining Observations and Models: A Review of the CARDAMOM Framework for Data‐Constrained Terrestrial Ecosystem Modeling

The rapid increase in the volume and variety of terrestrial biosphere observations (i.e., remote sensing data and in situ measurements) offers a unique opportunity to derive ecological insights, refine process‐based models, and improve forecasting for decision support. However, despite their potential, ecological observations have primarily been used to benchmark process‐based models, as many past and current models lack the capability to directly integrate observations and their associated uncertainties for parameterization. In contrast, data assimilation frameworks such as the CARbon DAta MOdel fraMework (CARDAMOM) and its suite of process‐based models, known as the Data Assimilation Linked Ecosystem Carbon Model (DALEC), are specifically designed for model‐data fusion. This review, motivated by a recent CARDAMOM community workshop, examines the development and applications of CARDAMOM, with an emphasis on its role in advancing ecosystem process understanding. CARDAMOM employs a Bayesian approach, using a Markov Chain Monte Carlo algorithm to enable data‐driven calibration of DALEC parameters and initial states (i.e., carbon pool sizes) through observation operators. CARDAMOM's unique ability to retrieve localized model process parameters from diverse datasets—ranging from in situ measurements to global satellite observations—makes it a highly flexible tool for analyzing spatially variable ecosystem responses to environmental change. However, assimilating these data also presents challenges, including data quality issues that propagate into model skill, as well as trade‐offs between model complexity, parameter equifinality, and predictive performance. We discuss potential solutions to these challenges, such as reducing parameter equifinality by incorporating new observations. This review also offers community recommendations for incorporating emerging datasets, integrating machine learning techniques, strengthening collaboration with remote sensing, field, and modeling communities, and expanding CARDAMOM's relevance for localized ecosystem monitoring and decision‐making. CARDAMOM enables a deep, mechanistic understanding of terrestrial ecosystem dynamics that cannot be achieved through empirical analyses of observational datasets or weakly constrained models alone.

Bayesian inference↗

A comparison of the power potential for surface- and seabed-deployed tidal turbines in the San Juan Archipelago, Salish Sea, WA

The San Juan archipelago lies along the axis of tidal movement between Straits of Juan de Fuca and Strait of Georgia in the Salish Sea. The amplitude of the tidal exchange produces significant tidal currents between the islands, as well as in Rosario and Haro Straits. These are of interest as a future source of electrical power generation, given the islands’ dependence on electricity supply by a subsea cable from the mainland. Here, we evaluate the tidal current energy potential in this region through a re-analysis of measurements collected by the National Ocean Service (NOS) and a high-resolution numerical model. Given the considerable variations in water depth and vertical velocity profiles across candidate tidal energy sites, we consider the trade-offs between tidal turbines deployed from a floating platform and those anchored to the seabed. Measurement re-analysis indicates several locations that could support tidal current power generation by MW-scale turbines with an acceptable balance between turbine size, rated power, and capacity factor. Even for relatively large (30 m) turbine diameters, surface-deployed turbines would be expected to produce up to 30 % more electricity than the same turbine deployed near the seabed due to vertical shear, with this difference increasing for smaller diameter turbines. A benchmark comparison at the measurement locations with the highest power generation potential shows relatively good model fidelity, though, even with relatively high resolution, time-average power density disagreements of ± 50 % persist throughout the water column. Nonetheless, the simulation identifies several locations with power generation potential more than twice as high as locations in the measurement re-analysis. These sites were either not surveyed by NOS or excluded due to data quality issues. Altogether, these results demonstrate the complimentary of models and measurements for tidal energy site assessment and suggest that tidal currents could be an important source of electricity generation in the San Juan archipelago.

16 TIDAL AND WAVE POWER↗

Detecting outliers in streaming time series data from ARM distributed sensors

The Atmospheric Radiation Measurement (ARM) Data Center at ORNL collects data from a number of permanent and mobile facilities around the globe. The data is then ingested to create high level scientific products. High frequency streaming measurements from sensors and radar instruments at ARM sites require high degree of accuracy to enable rigorous study of atmospheric processes. Outliers in collected data are common due to instrument failure or extreme weather events. Thus, it is critical to identify and flag them. We employed multiple univariate, multivariate and time series techniques for outlier detection methods and studied their effectiveness. First, we examined Pearson correlation coefficient which is used to measure the pairwise correlations between variables. Singular Spectrum Analysis (SSA) was applied to detect outliers by removing the anticipated annual and seasonal cycles from the signal to accentuate anomalies. K-means was applied for multivariate examination of data from collection of sensors to identify any deviation from expected and known patterns and identify abnormal observation. The Pearson correlation coefficient, SSA and K-means methods were later combined together in a framework to detect outliers through a range of checks. We applied the developed method to data from meteorological sensors at ARM Southern Great Plains site and validated against existing database of known data quality issues.

Lu, Yuping↗

Comparing evapotranspiration from Eddy covariance measurements, water budgets, remote sensing, and land surface models over Canada a, b

Here, this study compares six evapotranspiration ET products for Canada's landmass, namely, eddy covariance EC measurements; surface water budget ET; remote sensing ET from MODIS; and land surface model (LSM) ET from the Community Land Model (CLM), the Ecological Assimilation of Land and Climate Observations (EALCO) model, and the Variable Infiltration Capacity model (VIC). The ET climatology over the Canadian landmass is characterized and the advantages and limitations of the datasets are discussed. The EC measurements have limited spatial coverage, making it difficult for model validations at the national scale. Water budget ET has the largest uncertainty because of data quality issues with precipitation in mountainous regions and in the north. MODIS ET shows relatively large uncertainty in cold seasons and sparsely vegetated regions. The LSM products cover the entire landmass and exhibit small differences in ET among them. Annual ET from the LSMs ranges from small negative values to over 600 mm across the landmass, with a countrywide average of 256 ± 15 mm. Seasonally, the countrywide average monthly ET varies from a low of about 3 mm in four winter months (November-February) to 67 ± 7 mm in July. The ET uncertainty is scale dependent. Larger regions tend to have smaller uncertainties because of the offset of positive and negative biases within the region. More observation networks and better quality controls are critical to improving ET estimates. Future techniques should also consider a hybrid approach that integrates strengths of the various ET products to help reduce uncertainties in ET estimation.

54 ENVIRONMENTAL SCIENCES↗

Quality Data Essential for Modeling Water Cycles Effectively

The experimental and observational data is the backbone of effective modeling of all aspects of water cycles. As the computer technology used for these data gathering efforts has improved dramatically in the recent decades, the environmental conditions these instruments are subjected to remain pretty much the same. These measuring instruments are exposed to the sun light, wind, rain, snow and ice; and they are immersed in dirt, water and mud. All of these are damaging to the electronics components used for data gathering, storage, and transmission, which creates many different kinds of data quality issues that affect the quality and reliability of modeling and prediction efforts.

54 ENVIRONMENTAL SCIENCES↗

An Updated Life Cycle Assessment of Utility-Scale Solar Photovoltaic Systems Installed in the United States

Given the high deployment targets for solar photovoltaics (PV) needed to meet U.S. decarbonization goals, and the limited carbon budget remaining to limit global temperature rise, accurate accounting of the energy-use and greenhouse-gas emissions over the life-cycle of PV systems is needed. In the United States, most PV systems are large utility-scale systems which use single-axis trackers and central inverters, which are not commonly examined in existing life-cycle assessment (LCA) literature. In this study, we present a cradle-to-grave LCA of a typical silicon U.S. utility PV (UPV) installation which is consistent with the utility system features documented in the annual NREL PV system cost benchmark reports. We analyze and present results for four main metrics: cumulative energy demand (CED), greenhouse gas (GHG) emissions, energy payback time (EPBT), and carbon payback time (CPBT). We consider six primary manufacturing options: three based on an imported PV module supply chain (comparing low-carbon imports, high-carbon imports, and average imports), and three based on a potential domestic PV module supply chain (comparing low-carbon U.S. regions, high-carbon U.S. regions, and average U.S. regions). These manufacturing options were then paired with installation locations to create six main cases: low-carbon options were installed in Phoenix (Arizona), high-carbon options were installed in Seattle (Washington), and average options were installed in Fredonia (Kansas). These locations were selected to represent a range of irradiance and grid mixes in the United States, in order to illustrate the likely range of EPBTs and CPBTs possible across the United States. For all six cases, a sensitivity analysis for end-of-life (EOL) handling was explored to capture current and future management options: landfilling, partial recycling, and high-quality recycling. For the purposes of this report, the benchmark system was defined to use an average imported supply chain with partial recycling, installed in Fredonia (Kansas). CED results show ratios at or below 0.1 MJ oil-eq /MJgenerated which demonstrates efficient use of primary energy resources (below a 1:1 ratio), and represents a slight improvement over previous results in literature. GHG emissions per kWh range from 10-36 g CO 2 e, which are consistent with or lower than previous results published by NREL and IEA-PVPS. We use a graphical approach for calculating EPBT and CPBT in this report, which improves upon methods typically used in literature by accounting for non-linearity and avoiding data quality issues associated with long-term projections. EPBT was determined to vary from 0.5 to 1.2 years, with a benchmark EPBT of 0.6 years; CPBT was shown to vary from 0.8 to 20 years, with benchmark CPBT of 2.1 years, which is lower than other estimates from recent literature (typically >2 years).

14 SOLAR ENERGY↗

A Conceptual Site Model for Nature and Extent of Contamination in a Riparian-Near Shore Area - 12410

The 100-K decision area is located along the Columbia River and includes source Operable Units (OUs), a groundwater OU, and the adjacent surface water, saturated sediment and aquatic biota. A conceptual site model (CSM) has been developed to evaluate concentrations of non-radiological substances and radionuclides detected in soil, water and sediments in a riparian near-shore area along the Columbia River. The CSM is used to determine if potential transport pathways exist to these media from Hanford Site sources by incorporating information from the physical system, surface hydrology, subsurface hydrogeology, analytical results, and ecological evaluation into the model. Six contaminants of ecological concern, mostly metals and non-radiological inorganics, have been identified in riparian and near-shore media. With few exceptions (notably chromium and hexavalent chromium), there are ambient sources for these constituents in soil, sediment and water that are unrelated to the Hanford Site. While the CSM documented analytical and biological conditions, this paper presents results focused on analytical measurements to document the potential for these contaminants to be related to a release from the Hanford Site. The purpose for preparing this CSM was to address, on a reactor decision area basis, the potential for Hanford Site contaminants in soil or groundwater to migrate to riparian or near-shore areas at concentrations that could be of concern for ecological receptors. This CSM supplements the analysis of River Corridor-wide ecological risks presented in the ecological risk assessment of the RCBRA. The RCBRA identified on a site-wide basis some contaminants of ecological concern that warranted further evaluation. Based on the results of the further evaluation contained in this CSM, with the exception of hexavalent chromium, detected concentrations of contaminants in riparian or near-shore media are not reliably detectable at levels of ecological concern, or are not associated with contamination in soil or groundwater resulting from Hanford Site operations. Therefore, for purposes of alternatives evaluation in the 100-K Feasibility Study, hexavalent chromium in groundwater should be considered the only contaminants of ecological concern. There are uncertainties in this evaluation related to analytical data quality. With the exception of most recently collected analytical data in groundwater, analytical methods had MDLs that were either close to or higher than aquatic criteria; when methods with adequate detection limits are used, the results show that contaminant concentrations are less than aquatic criteria. In many cases where analytical results were higher than aquatic criteria, it was determined that the results were based on unfiltered samples. Unfiltered analytical results are inappropriate for comparison with aquatic criteria; in general, filtered analytical results were less than aquatic criteria. In some cases where concentrations were reported higher than aquatic criteria, further evaluation of the data revealed it was qualified because of the presence of blank contamination. The analytical detection limit considerations and data quality issues further support the conclusion that contaminants concentrations in water were not reliably detected above levels of ecological significance. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Robust Method to Identify Groundwater Affected By Redox Conditions at Los Alamos National Laboratory Legacy Cleanup Site - 20497

To facilitate the mission of the Department of Energy's (DoE's) Environmental Management Program at Los Alamos National Laboratory (LANL), a significant number of groundwater monitoring wells with depths ranging from 152-396 meter were completed for characterization and monitoring. Groundwater at Los Alamos tends to be oxygen saturated with very low concentrations of organic matter. Drilling fluids can introduce residual carbon causing an increase in microbial activity and local reducing conditions around the well, potentially impacting representativeness of groundwater data quality for redox sensitive constituents. We assert that the redox state of the wells can robustly be identified by aqueous solution concentrations of iron and manganese. An automated review process has been built into a computer-based data management system enabling an efficient screening process for identifying reducing conditions that are not caught using standard data-validation protocol, and ensures that important data quality issues are thoroughly identified and addressed in an efficient and timely manner. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Quality of Big Data in Healthcare

The current trend in Big Data Analytics and in particular Health information technology is towards building sophisticated models, methods and tools for business, operational and clinical intelligence, but the critical issue of data quality required for these models is not getting the attention it deserves. The objective of the paper is to highlight the issues of data quality in the context of Big Data Healthcare Analytics.

97 MATHEMATICS AND COMPUTING↗

Core Community Specifications for Electron Microprobe Operating Systems: Software, Quality Control, and Data Management Issues

Modem electron microprobe systems have become increasingly sophisticated. These systems utilize either UNIX or PC computer systems for measurement, automation, and data reduction. These systems have undergone major improvements in processing, storage, display, and communications, due to increased capabilities of hardware and software. Instrument specifications are typically utilized at the time of purchase and concentrate on hardware performance. The microanalysis community includes analysts, researchers, software developers, and manufacturers, who could benefit from exchange of ideas and the ultimate development of core community specifications (CCS) for hardware and software components of microprobe instrumentation and operating systems.

Fournelle, John↗

A Quality Screening Service for Remote Sensing Data

NASA provides a wide variety of Earth-observing satellite data products to a diverse community. These data are annotated with quality information in a variety of ways, with the result that many users struggle to understand how to properly account for quality when dealing with satellite data. To address this issue, a Data Quality Screening Service (DQSS) is being implemented for a number of datasets. The DQSS will enable users to obtain data files in which low-quality pixels have been filtered out, based either on quality criteria recommended by the science team or on the user s particular quality criteria. The objective is to increase proper utilization of this critical quality data in science data analysis of satellite data products.

Lynnes, Christopher↗

Perspectives on Citizen Science Data Quality

Information about data quality helps potential data users to determine whether and how data can be used and enables the analysis and interpretation of such data. Providing data quality information improves opportunities for data reuse by increasing the trustworthiness of the data. Recognizing the need for improving the quality of citizen science data, we describe quality assessment and quality control (QA/QC) issues for these data and offer perspectives on aspects of improving or ensuring citizen science data quality and for conducting research on related issues.

54 ENVIRONMENTAL SCIENCES↗

Improving Data and Prediction Quality of High-Throughput Perovskite Synthesis with Model Fusion

Combinatorial fusion analysis (CFA) is an approach for combining multiple scoring systems using the rank-score characteristic function and cognitive diversity measure. One example is to combine diverse machine learning models to achieve better prediction quality. In this work, we apply CFA to the synthesis of metal halide perovskites containing organic ammonium cations via inverse temperature crystallization. Using a data set generated by high-throughput experimentation, four individual models (support vector machines, random forests, weighted logistic classifier, and gradient boosted trees) were developed. We characterize each of these scoring systems and explore 66 possible combinations of the models. When measured by the precision on predicting crystal formation, the majority of the combination models improves the individual model results. The best combination models outperform the best individual models by 3.9 percentage points in precision. In addition to improving prediction quality, we demonstrate how the fusion models can be used to identify mislabeled input data and address issues of data quality. In particular, we identify example cases where all single models and all fusion models do not give the correct prediction. Experimental replication of these syntheses reveals that these compositions are sensitive to modest temperature variations across the different locations of the heating element that can hinder or enhance the crystallization process. In summary, we demonstrate that model fusion using CFA can not only identify a previously unconsidered influence on reaction outcome but also be used as a form of quality control for high-throughput experimentation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Supporting Global Air Quality Management Needs With A Flexible Data Fusion Tool for Estimation and Forecasting in Google Earth Engine

High spatial and temporal resolution air quality estimation and forecasting can be enhanced by combining global data sources, like chemical transport models and satellite remote sensing, with local information from regulatory and low-cost air quality monitors. Successful integration of data from these diverse sources is complicated by many factors, however, including differences in spatial and temporal resolution, data availability and latency issues, varying data quality, and large computational and data storage requirements. This presentation will provide an overview of a NASA-funded effort to develop the foundation for future operationalization of air quality forecasting for world-wide end-users and integration into their air quality management decision processes, which will be achieved in future phases of this multi-year project. We will summarize our progress in developing a data fusion system using the Google Earth Engine platform which can integrate model, satellite, and surface-level monitoring datasets to enhance estimation and forecasting of air-quality-relevant pollutants at sub-daily and sub-city scales. The tool is being developed in close cooperation with several city- and regional-level air quality managers in the USA and around the world. Our end-goal is to provide these air quality managers with the information they need to assess and anticipate the impacts of poor air quality, track changes in air quality due to ongoing mitigation efforts and land use changes, and identify ways to improve their air quality monitoring strategies. This presentation will focus on recent advances achieved through the project, including integration of multiple air quality datasets in a prototype data fusion system in Google Earth Engine, the quantification of uncertainties associated with our data fusion approach, and the development of user interfaces and visualization tools to convey air quality information in a way which best meets end-user needs.

Carl Malings↗

Supporting Global Air Quality Management Needs With A Flexible Data Fusion Tool for Estimation and Forecasting in Google Earth Engine

High spatial and temporal resolution air quality estimation and forecasting can be enhanced by combining global data sources, like chemical transport models and satellite remote sensing, with local information from regulatory and low-cost air quality monitors. Successful integration of data from these diverse sources is complicated by many factors, however, including differences in spatial and temporal resolution, data availability and latency issues, varying data quality, and large computational and data storage requirements. This presentation will provide an overview of a NASA-funded effort to develop the foundation for future operationalization of air quality forecasting for world-wide end-users and integration into their air quality management decision processes, which will be achieved in future phases of this multi-year project. We will summarize our progress in developing a data fusion system using the Google Earth Engine platform which can integrate model, satellite, and surface-level monitoring datasets to enhance estimation and forecasting of air-quality-relevant pollutants at sub-daily and sub-city scales. The tool is being developed in close cooperation with several city- and regional-level air quality managers in the USA and around the world. Our end-goal is to provide these air quality managers with the information they need to assess and anticipate the impacts of poor air quality, track changes in air quality due to ongoing mitigation efforts and land use changes, and identify ways to improve their air quality monitoring strategies. This presentation will focus on recent advances achieved through the project, including integration of multiple air quality datasets in a prototype data fusion system in Google Earth Engine, the quantification of uncertainties associated with our data fusion approach, and the development of user interfaces and visualization tools to convey air quality information in a way which best meets end-user needs.

Carl Malings↗

Increasing Efficiency at the NTF by Optimizing Model AoA Positioning

The National Transonic Facility (NTF) at NASA Langley Research Center (LaRC) is a national resource for aeronautical research and development. The government, military and private industries rely on the capability of this facility for realistic flight data. Reducing the operation costs and keeping the NTF affordable is essential for aeronautics research. The NTF is undertaking an effort to reduce the time between data points during a pitch polar. This reduction is being driven by the operating costs of a cryogenic facility. If the time per data point can be reduced, a substantial cost savings can be realized from a reduction in liquid nitrogen (LN2) consumption. It is known that angle-of-attack (AoA) positioning is the longest lead-time item between points. In January 2005 a test was conducted at the NTF to determine the cause of the long lead-time so that an effort could be made to improve efficiency. The AoA signal at the NTF originates from onboard instrumentation then travels through a number of different systems including the signal conditioner, digital voltmeter, and the data system where the AoA angle is calculated. It is then fed into a closed loop control system that sets the model position. Each process along this path adds to the time per data point affecting the efficiency of the data taking process. Due to the nature of the closed loop feed back AoA control and the signal path, it takes approximately 18 seconds to take one pitch pause point with a typical AoA increment. Options are being investigated to reduce the time delay between points by modifying the signal path. These options include: reduced signal filtering, using analog channels instead of a digital volt meter (DVM), re-routing the signal directly to the AoA control computer and implementing new control algorithms. Each of these has potential to reduce the positioning time and together the savings could be significant. These timesaving efforts are essential but must be weighed against possible loss of data quality. For example, a reduction in filtering can introduce noise into the signal and using analog channels could result in some loss of accuracy. Data quality assessments need to be performed concurrently with timesaving techniques since data quality parameters are essential in maintaining facility integrity. This paper will highlight time saving efforts being undertaken or studied at the NTF. It will outline the instrumentation and computer systems involved in setting of the model pitch attitude then suggest changes to the process and discuss how these system changes would effect the time between data points. It also discusses the issue of data quality and how the potential efficiency changes in the system could affect it. Lastly, it will discuss the possibility of using an open loop control system and give some pros and cons of this method.

Crawford, Bradley L.↗

Genomes to Proteomes

Biologists are awash with genomic sequence data. In large part, this is due to the rapid acceleration in the generation of DNA sequence that occurred as public and private research institutes raced to sequence the human genome. In parallel with the large human genome effort, mostly smaller genomes of other important model organisms were sequenced. Projects following on these initial efforts have made use of technological advances and the DNA sequencing infrastructure that was built for the human and other organism genome projects. As a result, the genome sequences of many organisms are available in high quality draft form. While in many ways this is good news, there are limitations to the biological insights that can be gleaned from DNA sequences alone; genome sequences offer only a bird's eye view of the biological processes endemic to an organism or community. Fortunately, the genome sequences now being produced at such a high rate can serve as the foundation for other global experimental platforms such as proteomics. Proteomic methods offer a snapshot of the proteins present at a point in time for a given biological sample. Current global proteomics methods combine enzymatic digestion, separations, mass spectrometry and database searching for peptide identification. One key aspect of proteomics is the prediction of peptide sequences from mass spectrometry data. Global proteomic analysis uses computational matching of experimental mass spectra with predicted spectra based on databases of gene models that are often generated computationally. Thus, the quality of gene models predicted from a genome sequence is crucial in the generation of high quality peptide identifications. Once peptides are identified they can be assigned to their parent protein. Proteins identified as expressed in a given experiment are most useful when compared to other expressed proteins in a larger biological context or biochemical pathway. In this chapter we will discuss the automatic annotation and the generation of high quality gene models, the setup and execution of global proteomics experiments that are quantitative and statistically rigorous and finally add biological context to proteomics.

59 BASIC BIOLOGICAL SCIENCES↗