Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Average Accuracy”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
NASA NTRS1995

Coral Pigments: Quantification Using HPLC and Detection by Remote Sensing

Widespread coral bleaching (loss of pigments of symbiotic dinoflagellates), and the corresponding decline in coral reef health worldwide, mandates the monitoring of coral pigmentation. Samples of the corals Porites compressa and P. lobata were collected from a healthy reef at Puako, Hawaii, and chlorophyll (chl) a, peridinin, and Beta-carotene (Beta-car) were quantified using reverse-phase high performance liquid chromatography (HPLC). Detailed procedures are presented for the extraction of the coral pigments in 90% acetone, and the separation, identification, and quantification of the major zooxanthellar pigments using spectrophotometry and a modification of the HPLC system described by Mantoura and Llewellyn (1983). Beta-apo-8-carotenal was found to be inadequate as in internal standard, due to coelution with chl b and/or chl a allomer in the sample extracts. Improvements are suggested, which may result in better resolution of the major pigments and greater accuracy in quantification. Average concentrations of peridinin, chl a, and Beta-car in corals on the reef were 5.01, 8.59, and 0.29, micro-grams/cm(exp 2), respectively. Average concentrations of peridinin and Beta-car did not differ significantly between the two coral species sampled; however, the mean chl a concentration in P. compressa specimens (7.81 ,micro-grams/cm(exp 2) was significantly lower than that in P. lobata specimens (9.96 11g/cm2). Chl a concentrations determined spectrophotometrically were significantly higher than those generated through HPLC, suggesting that spectrophotometry overestimates chl a concentrations. The average ratio of chl a-to-peridinin concentrations was 1.90, with a large (53%) coefficient of variation and a significant difference between the two species sampled. Additional data are needed before conclusions can be drawn regarding average pigment concentrations in healthy corals and the consistency of the chl a/peridinin ratio. The HPLC pigment concentration values contribute to the limited database of pigment concentrations in healthy corals, from which quantitative definitions of 'healthy' vs. 'bleached' coral may emerge. They also serve as ground-truth, corresponding to fluorescence data collected from the reef at Puako using airborne remote sensing of laser induced fluorescence. Fluorescence spectra from several overflights using the NASA AOL (airborne oceanographic lidar) system show consistent chlorphyll fluorescence peaks around 685 nm, as well as consistence peaks in the 400-600 nm range which may emanate from granules in the coral tissue. These data, along with results from previous studies of coral fluorescence, suggest that remote sensing of laser-induced fluorescence may become a rapid and efficient means of monitoring coral pigmentation and coral reef bleaching.

Cottone, Mary C.↗
NASA NTRS1999

Preliminary Global Topographic Model of Mars Based on MOLA Altimetry, Earth-Based Radar, and Viking, Mariner and MGS Occultations

The recent altimetry data acquired by MOLA over the northern hemisphere of Mars have been combined with the Earth-based radar data obtained between 1971 and 1982, and occultation measurements of the Viking 1 and 2 Orbiters, Mariner 9, and MGS to derive a global model of the shape and topography of Mars. This preliminary model has a horizontal resolution of about 300 km. Vertical accuracy is on average a few hundred meters in the region of the data. Datasets: The altimetry and radar datasets were individually binned in 1.25 degree grids and merged with the occultation data. The Viking and Mariner occultation data in the northern hemisphere were excluded from the combined dataset where MOLA altimetry were available. The laser altimetry provided extensive and almost complete coverage of the northern hemisphere north of latitude 30 while the radar provided longitudinal coverage at several latitude bands between 23N and 23S. South of this region the only data were occultations. The majority of the occultations were obtained from Mariner 9, and the rest from Viking 1 & 2, and MGS. Earlier studies had shown that the Viking and Mariner occultations were on average only accurate to 500 meters. The recent MGS occultations are accurate to a few tens of meters. However, the highest southern latitude reached by the MGS occultations is only about 64S and data near the target region for the Mars 98 lander is limited to a few Viking and Mariner observations of relatively poor quality. In addition to the above datasets the locations of the Viking 1, Viking 2, and Pathfinder landers, obtained from the radio tracking of their signals, were included.

Smith, David E.↗
NASA NTRS2000

Sensing Water Vapon via Spacecraft Radio Occultation Observations

The radio occultation technique has been used to characterize planetary atmospheres since the 1960's spanning atmospheric pressures from 16 microbars to several bars. In 1988, the use of GPS signals to make occultation observations of Earth's atmosphere was realized by Tom Yunck and Gunnar Lindal at JPL. In the GPS to low-Earth-orbiter limb- viewing occultation geometry, Fresnel diffraction yield a unique combination of high vertical resolution of 100 m to 1 km at long wavelengths (approx. 20 cm) insensitive to particulate scattering which allows routine limb sounding from the lower mesosphere through the troposphere. A single orbiting GPS/GLONASS receiver can observe - 1000 to 1400 daily occultations providing as many daily, high vertical resolution soundings as the present global radiosonde network, but with far more evenly distributed, global coverage. The occultations yield profiles of refractivity as a function of height. In the cold, dry conditions of the upper troposphere and above (T less than 240 K), profiles of density, pressure (geopotential), and temperature can be derived. Given additional temperature information, water vapor can be derived in the midddle and lower troposphere with a unique combination of vertical resolution, global distribution and insensitivity to clouds and precipitation to an accuracy of approx. 0.2 g/kg. At low latitudes, moisture profiles will be accurate to 1-5% within the convective boundary layer and better than 20% below 6 to 7 km. Accuracies of climatological averages should be approx. 0. 1 g/kg limited by the biases in the temperature estimates. To use refractivity to constrain water vapor, knowledge of temperature is required. The simplest approach is to use the temperature field from an analysis such as the 6 hour ECMWF global analysis interpolated to the locations of each occultation. A better approach is to combine the temperature and moisture fields from such an analysis with the occultation refractivity in a weighting scheme based on the errors in each data field. A ID variational combinational approach has been developed at the UKMO. We win present results from both approaches from GPS/MET data taken in June and July 1995 and compare them with the ECMWF global 6 hour moisture analyses which are derived largely from TOVS and radiosonde data. Overall, the atmosphere below the 500 mb level appears somewhat drier in general than the ECNIWF humidity field. A 2-D (latitude vs. height) climatological snapshot derived from a 2-week span of GPS/MET data will be compared to the humidity climatology of Peixoto and Oort derived from radiosonde data from 1963-1973. Differences between the GPS results and Peixoto and Oort may be the signature of a climate trend over the past 30 years.

Kursinski, E. Robert↗
NASA NTRS2000

TRMM Microwave Radiometer Rain Rate Estimation Method with Convective and Stratiform Discrimination

Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) radiometer brightness temperature data in the 85 GHz channel (T85) reveal distinct local minima (T85min) in a regional map containing a Mesoscale Convective System (MCS). A map of surface rain rate for that region, deduced from simultaneous measurements made by the Precipitation Radar (PR) on board the TRMM satellite, reveals that these T85min, produced by scattering, correspond to local PR rain maxima. Utilizing the PR rain rate map as a guide, we have developed a TMI algorithm to retrieve convective and stratiform rain. In this algorithm, two parameters are used to classify three kinds of thunderstorms (Cbs) based on the T85 data: a) the magnitude of scattering depression deduced from local T85mi, and b) the mean horizontal gradient of T85 around such minima. Initially, the algorithm is optimized or tuned utilizing the PR and TMI data of a few MCS events. The areal distribution of light (1-10 mm/hr), moderate (10-20 mm/hr), and intense (greater than or equal to 20 mm/hr) rain rates are retrieved on the average with an accuracy of about 15%. Taking advantage of this ability of our retrieval method, one could derive the latent heat input into the atmosphere over the 760 km wide swath of the TMI radiometer in the tropics.

Prabhakara, Cuddapah↗
NASA NTRS2004

New Reduced Two-Time Step Method for Calculating Combustion and Emission Rates of Jet-A and Methane Fuel With and Without Water Injection

A simplified kinetic scheme for Jet-A, and methane fuels with water injection was developed to be used in numerical combustion codes, such as the National Combustor Code (NCC) or even simple FORTRAN codes that are being developed at Glenn. The two time step method is either an initial time averaged value (step one) or an instantaneous value (step two). The switch is based on the water concentration in moles/cc of 1x10(exp -20). The results presented here results in a correlation that gives the chemical kinetic time as two separate functions. This two step method is used as opposed to a one step time averaged method previously developed to determine the chemical kinetic time with increased accuracy. The first time averaged step is used at the initial times for smaller water concentrations. This gives the average chemical kinetic time as a function of initial overall fuel air ratio, initial water to fuel mass ratio, temperature, and pressure. The second instantaneous step, to be used with higher water concentrations, gives the chemical kinetic time as a function of instantaneous fuel and water mole concentration, pressure and temperature (T4). The simple correlations would then be compared to the turbulent mixing times to determine the limiting properties of the reaction. The NASA Glenn GLSENS kinetics code calculates the reaction rates and rate constants for each species in a kinetic scheme for finite kinetic rates. These reaction rates were then used to calculate the necessary chemical kinetic times. Chemical kinetic time equations for fuel, carbon monoxide and NOx were obtained for Jet-A fuel and methane with and without water injection to water mass loadings of 2/1 water to fuel. A similar correlation was also developed using data from NASA's Chemical Equilibrium Applications (CEA) code to determine the equilibrium concentrations of carbon monoxide and nitrogen oxide as functions of overall equivalence ratio, water to fuel mass ratio, pressure and temperature (T3). The temperature of the gas entering the turbine (T4) was also correlated as a function of the initial combustor temperature (T3), equivalence ratio, water to fuel mass ratio, and pressure.

Molnar, Melissa↗
NASA NTRS2005

Simplified Two-Time Step Method for Calculating Combustion and Emission Rates of Jet-A and Methane Fuel With and Without Water Injection

A simplified kinetic scheme for Jet-A, and methane fuels with water injection was developed to be used in numerical combustion codes, such as the National Combustor Code (NCC) or even simple FORTRAN codes. The two time step method is either an initial time averaged value (step one) or an instantaneous value (step two). The switch is based on the water concentration in moles/cc of 1x10(exp -20). The results presented here results in a correlation that gives the chemical kinetic time as two separate functions. This two time step method is used as opposed to a one step time averaged method previously developed to determine the chemical kinetic time with increased accuracy. The first time averaged step is used at the initial times for smaller water concentrations. This gives the average chemical kinetic time as a function of initial overall fuel air ratio, initial water to fuel mass ratio, temperature, and pressure. The second instantaneous step, to be used with higher water concentrations, gives the chemical kinetic time as a function of instantaneous fuel and water mole concentration, pressure and temperature (T4). The simple correlations would then be compared to the turbulent mixing times to determine the limiting rates of the reaction. The NASA Glenn GLSENS kinetics code calculates the reaction rates and rate constants for each species in a kinetic scheme for finite kinetic rates. These reaction rates are used to calculate the necessary chemical kinetic times. Chemical kinetic time equations for fuel, carbon monoxide and NOx are obtained for Jet-A fuel and methane with and without water injection to water mass loadings of 2/1 water to fuel. A similar correlation was also developed using data from NASA's Chemical Equilibrium Applications (CEA) code to determine the equilibrium concentrations of carbon monoxide and nitrogen oxide as functions of overall equivalence ratio, water to fuel mass ratio, pressure and temperature (T3). The temperature of the gas entering the turbine (T4) was also correlated as a function of the initial combustor temperature (T3), equivalence ratio, water to fuel mass ratio, and pressure.

Molnar, Melissa↗
NASA NTRS2006

The Aquarius Ocean Salinity Mission High Stability L-band Radiometer

The NASA Earth Science System Pathfinder (ESSP) mission Aquarius, will measure global ocean surface salinity with approx.120 km spatial resolution every 7-days with an average monthly salinity accuracy of 0.2 psu (parts per thousand). This requires an L-band low-noise radiometer with the long-term calibration stability of less than or equal to 0.15 K over 7 days. The instrument utilizes a push-broom configuration which makes it impractical to use a traditional warm load and cold plate in front of the feedhorns. Therefore, to achieve the necessary performance Aquarius utilizes a Dicke radiometer with noise injection to perform a warm - hot calibration. The radiometer sequence between antenna, Dicke load, and noise diode has been optimized to maximize antenna observations and therefore minimize NEDT. This is possible due the ability to thermally control the radiometer electronics and front-end components to 0.1 Crms over 7 days.

Pellerano, Fernando A.↗
NASA NTRS2005

Ultra Stable Microwave Radiometers for Future Sea Surface Salinity Missions

The NASA Earth Science System Pathfinder (ESSP) mission Aquarius will measure global sea surface salinity with 100-km spatial resolution every 8 days with an average monthly salinity accuracy of 0.2 psu (parts per thousand). This requires an L-band low-noise radiometer with the long-term calibration stability of less than 0.1 K over 8 days. This three-year research program on ultra stable radiometers has addressed the radiometer requirements and configuration necessary to achieve this objective for Aquarius and future ocean salinity missions. The system configuration and component performance have been evaluated with radiometer testbeds at both JPL and GSFC. The research has addressed several areas including component characterization as a function of temperature, a procedure for the measurement and correction for radiometer system non-linearity, noise diode calibration versus temperature, low noise amplifier performance over voltage, and temperature control requirements to achieve the required stability. A breadboard radiometer, utilizing microstrip-based technologies, has been built to demonstrate this long-term stability. This report also presents the results of the radiometer test program, a detailed radiometer noise model, and details of the operational switching sequence optimization that can be used to achieve the low noise and stability requirements. Many of the results of this research have been incorporated into the Aquarius radiometer design and will allow this instrument to achieve its goals.

sea surface salinity↗
NASA NTRS2012

Effect of Spectrally Varying Albedo of Vegetation Surfaces on Shortwave Radiation Fluxes and Aerosol Direct Radiative Forcing

This study develops an algorithm for representing detailed spectral features of vegetation albedo based on Moderate Resolution Imaging Spectrometer (MODIS) observations at 7 discrete channels, referred to as the MODIS Enhanced Vegetation Albedo (MEVA) algorithm. The MEVA algorithm empirically fills spectral gaps around the vegetation red edge near 0.7 micrometers and vegetation water absorption features at 1.48 and 1.92 micrometers which cannot be adequately captured by the MODIS 7 channels. We then assess the effects of applying MEVA in comparison to four other traditional approaches to calculate solar fluxes and aerosol direct radiative forcing (DRF) at the top of atmosphere (TOA) based on the MODIS discrete reflectance bands. By comparing the DRF results obtained through the MEVA method with the results obtained through the other four traditional approaches, we show that filling the spectral gap of the MODIS measurements around 0.7 micrometers based on the general spectral behavior of healthy green vegetation leads to significant improvement in the instantaneous aerosol DRF at TOA (up to 3.02Wm(exp -2) difference or 48% fraction of the aerosol DRF, .6.28Wm(exp -2), calculated for high spectral resolution surface reflectance from 0.3 to 2.5 micrometers for deciduous vegetation surface). The corrections of the spectral gaps in the vegetation spectrum in the near infrared, again missed by the MODIS reflectances, also contributes to improving TOA DRF calculations but to a much lower extent (less than 0.27Wm(exp -2), or about 4% of the instantaneous DRF). Compared to traditional approaches, MEVA also improves the accuracy of the outgoing solar flux between 0.3 to 2.5 micrometers at TOA by over 60Wm(exp -2) (for aspen 3 surface) and aerosol DRF by over 10Wm(exp -2) (for dry grass). Specifically, for Amazon vegetation types, MEVA can improve the accuracy of daily averaged aerosol radiative forcing in the spectral range of 0.3 to 2.5 micrometers at equator at the equinox by 3.7Wm(exp -2). These improvements indicate that MEVA can contribute to regional climate studies over vegetated areas and can help to improve remote sensing-based studies of climate processes and climate change.

Zhu, L.↗
NASA NTRS2013

Tomographic Observation and Bedmapping of Glaciers in Western Greenland with IceBridge Sounding Radar

We produced the high resolution bedmaps of several glaciers in western Greenland from IceBridge Mission sounding radar data using tomographic sounding technique. The bedmaps cover 3 regions: Russell glaciers, Umanaq glaciers and Jakobshavn glaciers of western Greenland. The covered areas is about 20x40 km(sup 2) for Russell glaciers and 300x100 sq km, and 100x80 sq km for Jakobshavn glaciers. The ground resolution is 50 meters and the average ice thickness accuracy is 10 to 20 meters. There are some void areas within the swath of the tracks in the bedmaps where the ice thickness is not known. Tomographic observations of these void areas indicate that the surface and shallow sub-surface pockets, likely filled with water, are highly reflective and greatly weaken the radar signal and reduce the energy reaching and reflected from the ice sheet bottom.

IceBridge missions↗
NASA NTRS2006

The Aquarius ocean salinity mission high stability L-band radiometer

The NASA Earth Science System Pathfinder (ESSP) mission Aquarius, will measure global ocean surface salinity with ~120 km spatial resolution every 7-days with an average monthly salinity accuracy of 0.2 psu (parts per thousand) [1]. This requires an L-band low-noise radiometer with the long-term calibration stability of ≤0.15 K over 7 days. The instrument utilizes a push-broom configuration which makes it impractical to use a traditional warm load and cold plate in front of the feedhorns. Therefore, to achieve the necessary performance Aquarius utilizes a Dicke radiometer with noise injection to perform a warm – hot calibration. The radiometer sequence between antenna, Dicke load, and noise diode has been optimized to maximize antenna observations and therefore minimize NEDT. This is possible due the ability to thermally control the radiometer electronics and front-end components to 0.1 °Crms over 7 days.

Freedman, Adam↗
NASA NTRS2021

The Analysis of Using Satellite Soil Moisture Observations for Flood Detection, Evaluating Over the Thailand’s Great Flood of 2011

A flood monitoring and warning system provides critical information that can protect property and save lives. A basin-scale flood monitoring system requires an effective observation platform that offers extensive ground coverage of flood conditions, low latency, and high spatiotemporal resolution. While satellite imagery offers substantial spatial flood extent in detail due to its high spatial resolution, the coarse temporal resolution and cloud obstruction limit its near real-time application. Daily soil moisture data derived from satellite sensors at a scale of a few km can be used to monitor extreme wet surface conditions arising in flood occurrences. This study analyses the flood detection capabilities of several sources of soil moisture information, including the Soil Moisture and Ocean Salinity mission (SMOS), the Advanced Microwave Scanning Radiometer on EOS (AMSR-E), the Advanced SCATterometer (ASCAT) on MetOp, the Global Land Data Assimilation System (GLDAS), and the WaterGAP Global Hydrology Model (WGHM). In addition to soil moisture, the analysis includes measurements of surface reflectance from the Moderate Resolution Imaging Spectroradiometer (MODIS), precipitation measurements from the Tropical Rainfall Measuring Mission (TRMM), and terrestrial water storage estimates from the Gravity Recovery And Climate Experiment (GRACE) as proxies for flood inundations. The analysis was conducted over the Chao Phraya River Basin (CPB) in Thailand, where the Great Flood of 2011 led to one of the most significant economic losses in the country's history. Satellite-derived soil moisture exhibits a stronger correlation with the flood inundations than the precipitation, model-derived soil moisture, and terrestrial water storage data. SMOS soil moisture observation agrees best with the MODIS-derived flood extent/occurrence, both in terms of spatial distribution and timing, and providing approximated flood lead-time of one week or longer. A neural network constructed from SMOS and MODIS data is used to predict flood intensity/occurrence (given soil moisture input) with a predicted time window from eight days to thirty-two days. The short-term prediction (e.g., eight days) achieves the highest accuracy with an averaged recovery rate of approximately 60% (correlation coefficient). This study's results suggest a potential application of satellite soil moisture data in assisting flood monitoring and warning systems.

Satellite soil moisture↗
NASA NTRSDate not supplied

Transcriptomics-based Machine Learning Analysis Predicts Space-Exposed Murine Livers

Limited sample sizes, high data dimensionality, and sensitivity to technical and biological variability of next generation sequencing (NGS), has typically limited machine learning (ML) in space studies and further study of radiation effects. However, pooling smaller studies while addressing intra- and inter-study variabilities allows for ML predictive modeling. Here, integration methods were applied to whole transcriptome shotgun sequencing (RNAseq) data from 6 mouse liver GeneLab datasets (GLDS) with a total of 113 spaceflight and ground-control samples to determine top features relevant to spaceflight including the effect of radiation exposure. Data was normalized within each study, then merged and scaled across all datasets. Data dimensionality was reduced using a minimum redundancy maximum relevance (MRMR) methodology. The top MRMR features were used to predict spaceflight vs. ground-control samples using a Random Forest (RF) classifier with 5-fold cross validation (CV). The ML-based gene sets were further compared against differential gene expression results from individual GLDS. CV training using the top 100 MRMR genes show averages of 86% accuracy and 0.95 AUC value on the validation set over 5 folds (Figure 1A). Baseline set analysis on differentially expressed genes (DEGs) identified using padj ≤ 0.05 show 811 or 68 DEGs overlapping between at least 2 or 3 studies, respectively (Figure 1B). Over-representation analysis showed overlapping biological processes related to fatty acid and lipid metabolism. Set analysis between the MRMR features and the DEGs showed 60 or 8 genes overlapping with at least 1 or 2 studies, respectively. MRMR feature selection and ensemble ML methods (e.g. RF) improve performance relative to a Naïve Bayes classifier when NGS data sets are analyzed. A challenge of applying ML methods across heterogeneous NGS data is accounting for signal:noise ratio. Here, signal validation across studies was shown by intersecting sets between top MRMR genes and DEGs from RNASeq analysis. Non-intersecting sets introduce opportunity to explore spaceflight relevant genes and implementing ML methods across existing NGS datasets may overcome sample size limitations. ML coupled with existing analytical methods enhances understanding of disease by revealing common underlying pathways across datasets.

Machine Learning↗
DOE OSTI2024

Decentralized Microgrid Protection Through Relative Fault Direction Classification: Preprint

Protection in inverter-based resources (IBRs) dominated microgrids generally face significant challenges due to the low fault current and inconsistent fault behaviors from IBRs. Recently, machine learning-based approaches have attracted considerable attention to address these challenges. This paper introduces a novel decentralized protection strategy for microgrids. The proposed method decomposes the protection challenge into several distributed learning tasks, enabling individual relays to autonomously determine the direction of faults using a binary classification framework based on support vector machine (SVM) algorithms. Following the distributed fault direction estimation, classifier outcomes are shared among neighboring relays, facilitating a local decision-making process to ascertain the presence of faults within the neighborhood. Finally, a tripping signal is generated based on the classifier results of each relay to operate the circuit breaker. To test and validate this approach, a 100% renewable microgrid model is simulated in MATLAB/Simulink. In the numerical analysis, the application of SVM classifiers in our approach yields impressive results: an average relay classification accuracy of 98%, and a 96% accuracy in circuit breaker control. These findings highlight the potential of machine-learning-based approaches in enhancing the efficiency and reliability of microgrid protection systems.

decentralized algorithm↗
NASA NTRS1994

Effect of radiance-to-reflectance transformation and atmosphere removal on maximum likelihood classification accuracy of high-dimensional remote sensing data

Many analysis algorithms for high-dimensional remote sensing data require that the remotely sensed radiance spectra be transformed to approximate reflectance to allow comparison with a library of laboratory reflectance spectra. In maximum likelihood classification, however, the remotely sensed spectra are compared to training samples, thus a transformation to reflectance may or may not be helpful. The effect of several radiance-to-reflectance transformations on maximum likelihood classification accuracy is investigated in this paper. We show that the empirical line approach, LOWTRAN7, flat-field correction, single spectrum method, and internal average reflectance are all non-singular affine transformations, and that non-singular affine transformations have no effect on discriminant analysis feature extraction and maximum likelihood classification accuracy. (An affine transformation is a linear transformation with an optional offset.) Since the Atmosphere Removal Program (ATREM) and the log residue method are not affine transformations, experiments with Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) data were conducted to determine the effect of these transformations on maximum likelihood classification accuracy. The average classification accuracy of the data transformed by ATREM and the log residue method was slightly less than the accuracy of the original radiance data. Since the radiance-to-reflectance transformations allow direct comparison of remotely sensed spectra with laboratory reflectance spectra, they can be quite useful in labeling the training samples required by maximum likelihood classification, but these transformations have only a slight effect or no effect at all on discriminant analysis and maximum likelihood classification accuracy.

Hoffbeck, Joseph P.↗
NASA NTRSDate not supplied

Improving Computational Efficiency of Prognostics Algorithms in Resource-Constrained Settings

In engineering and aerospace applications, it is vital to operational success to have insight into the expected performance and health of physical systems. The field of prognostics and health management provides quantitative methods for monitoring, predicting, and managing system health. Prognostics algorithms can be employed to assess the current state of a system, propagate the system throughout time, and predict potential anomalies or failures that may occur. While they can provide accurate prediction results, effective prognostics algorithms can be challenging to use in resource-constrained settings due to computational limitations and high computational latency, leading to obsolete predictions. Thus, computationally efficient and accurate algorithms are necessary for future remaining useful life predictions. In this work, we implement new algorithmic approaches for prediction, quantitatively compare them via a battery degradation use-case, and provide recommendations of potential improvements to a prognostics framework. One approach to prediction is through sampling, whereby the current state of a physical system is sampled many times and each sample is propagated forward until failure is reached, resulting in a distribution of failure values. To improve the efficiency of this process, we implemented five new algorithmic approaches to prediction, including three distinct sampling methods (standard Monte Carlo, Quasi-Monte Carlo, and Latin Hypercube Sampling), a variable time step algorithm, and a variable sample size algorithm. To compare the algorithms, we employ a variety of metrics designed specifically to analyze both computational efficiency and model accuracy. Our metrics include accuracy to compare the average predicted value to ground truth, mean absolute deviation to illustrate dispersion, specific percentile error to describe accuracy within a user-defined risk tolerance, and code run-time. To quantitatively analyze our results, we employ a use-case of degradation of a Lithium-ion battery. We use an electrochemistry-based model to describe the current health state of the battery, and implement our prediction algorithms to propagate forward in time until end-of-discharge (EOD) is reached. Notably, through this work it was found that none of our sampling approaches had a significant impact on computational efficiency or model accuracy in predicting EOD of the battery. We find that while the sampling methods are unique, the distributions they generate are similar, ultimately producing final predictions that are nearly identical. In exploring the effect of the time step within the prediction algorithm, we found that prediction accuracy was highly dependent on the time step used, and that implementing a variable time step within a particular prediction may provide an increase in computational efficiency while also maintaining prediction accuracy. Finally, implementing a variable sample size also affected prediction, and our results show that tuning both the magnitude and timing of the sample size adjustment can result in improved computation speed and maintained prediction accuracy. Taken together, our findings highlight the challenge of performing prognostics in resource-constrained settings, and illustrate the potential of developing new prediction algorithms to improve computational efficiency.

prognostics↗
NASA NTRS1991

Multi-Dimensional ENO Schemes for General Geometries

A class of ENO schemes is presented for the numerical solution of multidimensional hyperbolic systems of conservation laws in structured and unstructured grids. This is a class of shock-capturing schemes which are designed to compute cell-averages to high order accuracy. The ENO scheme is composed of a piecewise-polynomial reconstruction of the solution form its given cell-averages, approximate evolution of the resulting initial value problem, and averaging of this approximate solution over each cell. The reconstruction algorithm is based on an adaptive selection of stencil for each cell so as to avoid spurious oscillations near discontinuities while achieving high order of accuracy away from them.

Harten, Ami↗