SEARCH · Search NASA

Results for “Bad data”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 145 records · Page 8

All sources NASA NTRS DOE OSTI

NASA NTRS1992

Morning-evening differences in global and regional oceanic precipitation as observed by the SSM/I

For the present preliminary analysis of oceanic rainfall statistics, global oceanic SSM/I data were simply scanned for pixels which exhibited a 37 GHz polarization difference (vertically polarized brightness temperatures minus horizontally polarized brightness temperatures) of less than 15 K. Such a low polarization difference over the open ocean is a completely unambiguous indication of moderate to intense precipitation. Co-located brightness temperatures from all seven channels of the SSM/I were saved for each pixel so identified. Bad scans and geographically mislocated block of data were objectively identified and removed from the resulting data base. We collected global oceanic rainfall data for two time periods, each one month in length. The first period (20 July-19 August 1987) coincides with the peak of the Northern Hemisphere summer. The second period (13 January-12 February 1988) coincides with the Northern Hemisphere winter.

Petty, Grant W.↗

NASA NTRS1992

Morning-evening differences in global and regional oceanic precipitation as observed by the SSM/I

Petty, Grant W.↗

DOE OSTI2024

Deep Learning and Likelihood Approaches for Viral Phylogeography Converge on the Same Answers Whether the Inference Model Is Right or Wrong

Abstract Analysis of phylogenetic trees has become an essential tool in epidemiology. Likelihood-based methods fit models to phylogenies to draw inferences about the phylodynamics and history of viral transmission. However, these methods are often computationally expensive, which limits the complexity and realism of phylodynamic models and makes them ill-suited for informing policy decisions in real-time during rapidly developing outbreaks. Likelihood-free methods using deep learning are pushing the boundaries of inference beyond these constraints. In this paper, we extend, compare, and contrast a recently developed deep learning method for likelihood-free inference from trees. We trained multiple deep neural networks using phylogenies from simulated outbreaks that spread among 5 locations and found they achieve close to the same levels of accuracy as Bayesian inference under the true simulation model. We compared robustness to model misspecification of a trained neural network to that of a Bayesian method. We found that both models had comparable performance, converging on similar biases. We also implemented a method of uncertainty quantification called conformalized quantile regression that we demonstrate has similar patterns of sensitivity to model misspecification as Bayesian highest posterior density (HPD) and greatly overlap with HPDs, but have lower precision (more conservative). Finally, we trained and tested a neural network against phylogeographic data from a recent study of the SARS-Cov-2 pandemic in Europe and obtained similar estimates of region-specific epidemiological parameters and the location of the common ancestor in Europe. Along with being as accurate and robust as likelihood-based methods, our trained neural networks are on average over 3 orders of magnitude faster after training. Our results support the notion that neural networks can be trained with simulated data to accurately mimic the good and bad statistical properties of the likelihood functions of generative phylogenetic models.

Evolutionary Biology↗

DOE OSTI2021

Pink-beam serial femtosecond crystallography for accurate structure factor determination at an X-ray free-electron laser

Diffraction images with crystal hits from the large bandwidth (2.2 % BW) data set (thaumatin 6), and nominal SASE (0.17 % BW) data set (thaumatin 8 + thaumatin 10) with bad pixel masks, online indexing results and CrystFEL stream files from final processing.

Alvra↗

DOE OSTI2022

Graph Embeddings for CEBAF Operations: Progress and Future Plans

We describe research towards leveraging deep learning on graph representations of the injector beamline at the Continuous Electron Beam Accelerator Facility (CEBAF) in order to create a tool for improving the efficiency of beam tuning tasks. Specifically, we use graphs to represent the injector beamline at any arbitrary date and time and invoke a graph neural network to extract a low-dimensional, informative representation that can be visualized in two-dimensions. By analyzing years of operational data from the CEBAF archiver, good and bad regions of parameter space can be identified. The goal is to exercise this framework as a real-time tool to aid beam tuning, which represents the dominant source of machine downtime.

Tennant, C.↗

DOE OSTI2024

Graph Analytics for CEBAF Operations

We report on the progress achieved during a 2-year Laboratory Directed Research and Development (LDRD) project titled “Graph Analytics for CEBAF Operations”. The objective of this project is to leverage deep learning on graph representations of CEBAF’s injector beamline in order to create a tool for improving the efficiency of beam tuning tasks. Specifically, we use graphs to represent the injector beamline at any arbitrary date and time and invoke a graph neural network (GNN) to extract a low-dimensional, informative representation that can be visualized in two-dimensions. By analyzing years of operational data from the CEBAF archiver, good and bad regions of parameter space can be identified. The goal is to exercise this framework as a real-time tool to aid beam tuning, which represents the dominant source of machine downtime.

43 PARTICLE ACCELERATORS↗

NASA NTRS1983

Experimental deformation of polycrystalline H2O ice at high pressure and low temperature - Preliminary results

A preliminary study is carried out of involving 70 constant strain deformation tests on pure polycrystalline H2O ice under conditions covering most of the stability field of ice I sub h. Brittle failure of I sub h is found to be promoted by lower P, lower T, and higher strain rates. Ductile flow is found to be promoted by higher P, higher T, and lower strain rates. The brittle failure of ice I sub h is found to be most unusual. The fracture strength is a positive function of P only below 50 MPa. At pressures greater than this, the fracture strength is independent of P, and the fracture plane lies approximately 45 deg from the load axis. It is believed that existing extrapolation based on existing experimental data to Ganymede and Callisto may be badly in error.

Durham, W. B.↗

DOE OSTI2019

Characterizing and Improving the Data Reduction Pipeline for the Keck OSIRIS Integral Field Spectrograph

OSIRIS is a near-infrared (1.0–2.4 μm) integral field spectrograph operating behind the adaptive optics system at Keck Observatory and one of the first lenslet-based integral field spectrographs. Since its commissioning in 2005, it has been a productive instrument, producing nearly half the laser guide star adaptive optics papers on Keck. The complexity of its raw data format necessitated a custom data reduction pipeline (DRP) delivered with the instrument in order to iteratively assign flux in overlapping spectra to the proper spatial and spectral locations in a data cube. Other than bug fixes and updates required for hardware upgrades, the bulk of the DRP has not been updated since initial instrument commissioning. We report on the first major comprehensive characterization of the DRP using on-sky and calibration data. We also detail improvements to the DRP, including characterization of the flux assignment algorithm, exploration of spatial rippling in the reduced data cubes, and improvements to several calibration files, including the rectification matrix, bad-pixel mask, and wavelength solution. We present lessons learned from over a decade of OSIRIS data reduction that are relevant to the next generation of integral field spectrograph hardware and data reduction software design.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2024

On the Use of Mahalanobis Distance in Particle Image Velocimetry Post-Processing

Particle Image Velocimetry (PIV) is a method of flow measurement that has become increasingly popular as an experimental tool. New technology has made high-speed and higher-dimension (stereoscopic, tomographic, etc) methods available to an ever-growing population of researchers. These advanced methods can provide significantly more data than traditional low-speed planar PIV, but these larger data sets also require more resources to process and store. A major time sink in the post-processing of large PIV data sets is the identification and rejection of “bad” or ”spurious” vectors that survive an initial processing step in commercial software, which attempts to identify spurious vectors on an image-by-image basis and not with respect to repeated trials. The Mahalanobis distance, an almost 100-year-old statistical function, was determined to be well suited for this task for its computational efficiency and higher-dimensional nature. This work includes a summary of the Mahalanobis distance and validation of its usefulness as a tool for outlier rejection in the post-processing of PIV data.

Experimental Methods↗

NASA NTRS2005

Molecular Hydrogen Fluorescence in IC 63

This grant has supported the acquisition, reduction and analysis of data targeting the structure and excitation of molecular hydrogen in the reflection nebula IC 63 and in particular the fluorescent emission seen in the UV. In addition to manpower for analyzing the FUSE data, the grant supported the (attempted) acquisition of supporting ground-based data. We proposed for and received observing time for two sets of ground based, data; narrow band imaging ([S II], [O III) at KPNO (July 2002; Observer: Burgh) and imaging spectro-photometry of several of the near-infrared rotation-vibration lines of H2 at the IRTF (October 2003; Observer: Andersson). Unfortunately, both of these runs were failures, primarily because of bad weather, and did not result in any useful data. We combined the FUSE observations with rocket borne observations of the star responsible for exciting the H2 fluorescence in IC 63: gamma Cas, and with archival HUT observations of IC 63, covering the long-wavelength part of the molecular hydrogen fluorescence.

Andersson, B-G↗

NASA NTRS1985

The 4He(p,d)3He reaction at 200 and 400 MeV

The distorted wave calculations are compared with the He-4(p,d) cross section. At all four energies the theoretical description of the data is relatively poor. While the general magnitude of the cross sections is not in bad agreement, the detailed slopes of the theoretical angular distributions miss the shapes of the experimental data in a rather fundamental way.

Alons, P. W. F.↗

DOE OSTI2024

Graph learning for particle accelerator operations

Particle accelerators play a crucial role in scientific research, enabling the study of fundamental physics and materials science, as well as having important medical applications. This study proposes a novel graph learning approach to classify operational beamline configurations as good or bad. By considering the relationships among beamline elements, we transform data from components into a heterogeneous graph. We propose to learn from historical, unlabeled data via our self-supervised training strategy along with fine-tuning on a smaller, labeled dataset. Additionally, we extract a low-dimensional representation from each configuration that can be visualized in two dimensions. Leveraging our ability for classification, we map out regions of the low-dimensional latent space characterized by good and bad configurations, which in turn can provide valuable feedback to operators. This research demonstrates a paradigm shift in how complex, many-dimensional data from beamlines can be analyzed and leveraged for accelerator operations.

43 PARTICLE ACCELERATORS↗

NASA NTRS2004

RGS Spectroscopy of the Cygnus Loop XA Knot

Mushotzky, Richard F.↗

NASA NTRS2004

Formal Verification of Safety Properties for Aerospace Systems Through Algorithms Based on Exhaustive State-Space Exploration

The Runway Safety Monitor (RSM) designed by Lockheed Martin is part of NASA's effort to reduce aviation accidents. We developed a Petri net model of the RSM protocol and used the model checking functions of our tool SMART to investigate a number of safety properties in RSM. To mitigate the impact of state-space explosion, we built a highly discretized model of the system, obtained by partitioning the monitored runway zone into a grid of smaller volumes and by considering scenarios involving only two aircraft. The model also assumes that there are no communication failures, such as bad input from radar or lack of incoming data, thus it relies on a consistent view of reality by all participants. In spite of these simplifications, we were able to expose potential problems in the RSM conceptual design. Our findings were forwarded to the design engineers, who undertook corrective action. Additionally, the results stress the efficiency attained by the new model checking algorithms implemented in SMART, and demonstrate their applicability to real-world systems. Attempts to verify RSM with NuSMV and SPIN have failed due to excessive memory consumption.

Ciardo, Gianfranco↗

NASA NTRS2006

Formal Verification of the Runway Safety Monitor

The Runway Safety Monitor (RSM) designed by Lockheed Martin is part of NASA's effort to reduce runway accidents. We developed a Petri net model of the RSM protocol and used the model checking functions of our tool SMART to investigate a number of safety properties in RSM. To mitigate the impact of state-space explosion, we built a highly discretized model of the system, obtained by partitioning the monitored runway zone into a grid of smaller volumes and by considering scenarios involving only two aircraft. The model also assumes that there are no communication failures, such as bad input from radar or lack of incoming data, thus it relies on a consistent view of reality by all participants. In spite of these simplifications, we were able to expose potential problems in the RSM conceptual design. Our findings were forwarded to the design engineers, who undertook corrective action. Additionally, the results stress the efficiency attained by the new model checking algorithms implemented in SMART, and demonstrate their applicability to real-world systems.

Siminiceanu, Radu↗

NASA NTRS2003

RGS Spectroscopy of the Cygnus Loop XA Knot

The observations were performed at the end of April 2002, and the data were received in July 2002. Unfortunately, the observations were badly compromised by high levels of background radiation; one the three observations lost entirely. Two replacement observations were scheduled for November 2002, and were only made available in January of 2003. Consequently, we have had little time to grapple with the unusual data analysis challenges. The search for a postdoctoral fellow has been successfully concluded, and Manami Sasaki began working for us in January 2003. She will be supported in part by these funds, and will be working to help understand these data. Examination of the RGS 'Orders' images indicate the presence of broad emission lines (as expected for the diffuse XA knot). However, examination of the 'Spatial' dispersion/cross-dispersion images indicate that the emission is also broad in the cross-dispersion direction. (As a crosscheck, some of the 'Lockman Hole' datasets were also examined as representative 'sky background' datasets; in these, both types of images are relatively flat (outside the calibration source regions). The quicklook plots of the spectra show the expected O VII and O VIII lines, in addition to a complex around 35 Angstroms; the approx. 35 Angstrom line is likely the C V He-beta line at 34.97 Angstrom, but identifying the additional line(s) will require a more careful reduction of the data. Consequently, there is valuable information to be extracted from these data, but it is complicated by diffuse nature of the emission. Because the angular scale is large, we will have to make use of sky background datasets in order to do the background fitting. A color composite image of OM data in the three UV bands was presented at the 'How does the Galaxy Work?' meeting, and compared to optical and X-ray imaging data. Quantitative analysis will require obtaining the effective bandpasses of the UV filters so that the predominant line and continuum contributions can be identified using plasma shock emission models. In view of the facts that the data were only obtained relatively recently, and the complexity of the data analysis, we request a one year no-cost extension on the grant.

Gaetz, Terrance J.↗

NASA NTRS1993

Climatic variation of storms

Long-term variation of U.S. tornadoes were studied by obtaining the best possible data during the 75 years, 1916-90. The most difficult task was to estimate the number of early tornadoes in 1916-50 when the reporting efficiency was very bad, resulting in undercounting of the incidents. In generating the best possible data, Fujita's Tornado Tape produced at the University of Chicago, the book Significant Tornadoes, 1880-1989, by Thomas P. Grazulis, and the NSSFC Tornado Tape were combined. First, the annual number of tornadoes were smoothed to obtain the smoothed number of annual tornadoes which was normalized to the standard number of tornadoes defined as the mean number in the recent 30 years 1960-89. The normalization factor was obtained by computing the ration of smoothed and standard number. In the early years, the factor was in excess of 7. Thereafter, the normalized number of annual tornadoes was computed and plotted, finding that the results are very satisfactory. In order to visualize the variation of tornado activities, the tornado activity number (TAN) was initiated and computed. The TAN including all annual tornadoes show certain periodicity. For climatological evaluation, the peak-activity day of each year, such as Superoutbreak day, 3 April 1974, Palm Sunday, 11 April 1965 were eliminated. The TAN, excluding peak-day tornadoes, is of extreme interest.

Fujita, Ted↗

DOE OSTI2024

Newton-Raphson AC Power Flow Convergence Based on Deep Learning Initialization and Homotopy Continuation

Power flow forms the basis of many power system studies. With the increased penetration of renewable energy, grid planners tend to perform multiple power flow simulations under various operating conditions and not just selected snapshots at peak or light load conditions. Getting a converged AC power flow (ACPF) case remains a significant challenge for grid planners especially in large power grid networks. This paper proposes a two-stage approach to improve Newton-Raphson ACPF convergence and was applied to a 6102 bus Electric Reliability Council of Texas (ERCOT) system. The first stage utilizes a deep learning-based initializer with data re-training. Here a deep neural network (DNN) initializer is developed to provide better initial voltage magnitude and angle guesses to aid in power flow convergence. This is because Newton-Raphson ACPF is quite sensitive to the initial conditions and bad initialization could lead to divergence. The DNN initializer includes a data re-training framework that improves the initializer's performance when faced with limited training data. The DNN initializer successfully solved 3,285 cases out of 3,899 non-converging dispatch and performed better than random forest and DC power flow initialization methods. ACPF cases not solved in this first stage are then passed through a hot-starting algorithm based on homotopy continuation with switched shunt control. The hot-starting algorithm successfully converged 416 cases out of the remaining 614 non-converging ACPF dispatch. In conclusion, the combined two-stage approach achieved a 94.9% success rate, by converging a total of 3,701 cases out of the initial 3,899 unsolved cases.

Deep learning↗