SEARCH · Search NASA
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.
We describe progress towards the micron-scale analysis of krypton isotopes in extraterrestrial material.
Size distribution sampling errors of point plane electrostatic precipitator for aerosol sampling
Tests of arcing and current collection in simulated space plasma conditions have been performed at the NASA Glenn Research Center (GRC) in Cleveland, Ohio, for over 30 years and at the Marshall Space Flight Center (MSFC) in Huntsville, Alabama, for almost as long. During this period, proper test conditions for accurate and meaningful space simulation have been worked out, comparisons with actual space performance in spaceflight tests and with real operational satellites have been made, and NASA has achieved our own internal standards for test protocols. It is the purpose of this paper to communicate the test conditions, test procedures, and types of analysis used at NASA GRC and MSFC to the space environmental testing community at large, to help with international space-plasma arcing-testing standardization. Discussed herein are neutral gas conditions, plasma densities and uniformity, vacuum chamber sizes, sample sizes and Debye lengths, biasing samples versus self-generated voltages, floating samples versus grounded samples, test electrical conditions, arc detection, preventing sustained discharges during testing, real samples versus idealized samples, validity of LEO tests for GEO samples, extracting arc threshold information from arc rate versus voltage tests, snapover, current collection, and glows at positive sample bias, Kapton pyrolysis, thresholds for trigger arcs, sustained arcs, dielectric breakdown and Paschen discharge, tether arcing and testing in very dense plasmas (i.e. thruster plumes), arc mitigation strategies, charging mitigation strategies, models, and analysis of test results. Finally, the necessity of testing will be emphasized, not to the exclusion of modeling, but as part of a complete strategy for determining when and if arcs will occur, and preventing them from occurring in space.
In this paper, we consider a strongly convex stochastic optimization problem and propose three classes of variable sample-size stochastic first-order methods: (i) the standard stochastic gradient descent method, (ii) its accelerated variant, and (iii) the stochastic heavy-ball method. In each scheme, the exact gradients are approximated by averaging across an increasing batch size of sampled gradients. We prove that when the sample size increases at a geometric rate, the generated estimates converge in mean to the optimal solution at an analogous geometric rate for schemes (i)–(iii). Based on this result, we provide central limit statements, whereby it is shown that the rescaled estimation errors converge in distribution to a normal distribution with the associated covariance matrix dependent on the Hessian matrix, the covariance of the gradient noise, and the step length. If the sample size increases at a polynomial rate, we show that the estimation errors decay at a corresponding polynomial rate and establish the associated central limit theorems (CLTs). Under certain conditions, we discuss how both the algorithms and the associated limit theorems may be extended to constrained and nonsmooth regimes. As a result, we provide an avenue to construct confidence regions for the optimal solution based on the established CLTs and test the theoretical findings on a stochastic parameter estimation problem.
This document reports the status of the NEPP Double Data Rate 2 (DDR2) Reliability effort for FY2012. The task expanded the focus of evaluating reliability effects targeted for device examination. FY11 work highlighted the need to test many more parts and to examine more operating conditions, in order to provide useful recommendations for NASA users of these devices. In order to develop these approaches, it is necessary to develop test capability that can identify reliability outliers. To do this we must test many devices to ensure outliers are in the sample, and we must develop characterization capability to measure many different parameters. For FY12 we increased capability for reliability characterization and sample size. We increased sample size by moving from loose devices to DIMMs with an approximate reduction of 20 to 50 times in terms of per DUT cost. By increasing sample size we have improved our ability to characterize devices that may be considered reliability outliers. This report provides an update on the effort to improve DDR2 testing capability. Although focused on DDR2, the methods being used can be extended to DDR and DDR3 with relative ease.
This document reports the status of the NASA Electronic Parts and Packaging (NEPP) Double Data Rate 2 (DDR2) Reliability effort for FY2012. The task expanded the focus of evaluating reliability effects targeted for device examination. FY11 work highlighted the need to test many more parts and to examine more operating conditions, in order to provide useful recommendations for NASA users of these devices. This year's efforts focused on development of test capabilities, particularly focusing on those that can be used to determine overall lot quality and identify outlier devices, and test methods that can be employed on components for flight use. Flight acceptance of components potentially includes considerable time for up-screening (though this time may not currently be used for much reliability testing). Manufacturers are much more knowledgeable about the relevant reliability mechanisms for each of their devices. We are not in a position to know what the appropriate reliability tests are for any given device, so although reliability testing could be focused for a given device, we are forced to perform a large campaign of reliability tests to identify devices with degraded reliability. With the available up-screening time for NASA parts, it is possible to run many device performance studies. This includes verification of basic datasheet characteristics. Furthermore, it is possible to perform significant pattern sensitivity studies. By doing these studies we can establish higher reliability of flight components. In order to develop these approaches, it is necessary to develop test capability that can identify reliability outliers. To do this we must test many devices to ensure outliers are in the sample, and we must develop characterization capability to measure many different parameters. For FY12 we increased capability for reliability characterization and sample size. We increased sample size this year by moving from loose devices to dual inline memory modules (DIMMs) with an approximate reduction of 20 to 50 times in terms of per device under test (DUT) cost. By increasing sample size we have improved our ability to characterize devices that may be considered reliability outliers. This report provides an update on the effort to improve DDR2 testing capability. Although focused on DDR2, the methods being used can be extended to DDR and DDR3 with relative ease.
Mechanical testing campaigns are required to qualify materials for advanced reactor conditions, yet economical and safety limitations restrict the number of standardized mechanical tests that can be performed. Reducing the size of the sample is one approach to addressing these challenges and to accelerating testing. Previous research has shown that smaller mechanical test samples produce higher yield and ultimate stress values compared to values measured from standard sample sizes: the “smaller is stronger” effect. Specimens used in accelerated material testing campaigns must reflect bulk material performance to enable engineering scale material property measurement. The objective of this research project was to determine if engineering scale mechanical behavior—the yield stress—could be measured with micro-tensile test samples smaller than traditional standard testing geometries. The relationship between yield stress and sample size was explored with two different nuclear-relevant structural materials: Zircaloy-4 and tungsten. Mechanical testing of both metals demonstrated decreasing yield stress values with increasing sample gauge size across three different sizes. Yield stress values from the largest gauge size, fabricated with a femto-second laser ablation system, approach bulk material yield stress values reported in published literature. Preliminary analysis of the tungsten samples indicates the yield stress value depends on the grain characteristics within the gauge section, in addition to the gauge size. Accompanying modeling efforts, including response surface generation and crystal plasticity approaches, further demonstrated that the size of the sample gauge section alone cannot explain the change in yield stress values.
A study is made of the extent to which the size of the sample affects the accuracy of a quadratic or a cubic polynomial approximation of an experimentally observed quantity, and the trend with regard to improvement in the accuracy of the approximation as a function of sample size is established. The task is made possible through a simulated analysis carried out by the Monte Carlo method in which data are simulated by using several transcendental or algebraic functions as models. Contaminated data of varying amounts are fitted to either quadratic or cubic polynomials, and the behavior of the mean-squared error of the residual variance is determined as a function of sample size. Results indicate that the effect of the size of the sample is significant only for relatively small sizes and diminishes drastically for moderate and large amounts of experimental data.
A study was conducted on INCONEL® alloy 740H® to examine the role of grain size and sheet thickness on the alloy’s creep-rupture behavior. Three different starting sheet thicknesses were utilized and multiple heat-treatment conditions anticipated for compact heat exchanger (CHX) manufacturing were applied to produce a range of grain sizes. Creep-rupture testing was conducted at 750 °C for times up to about 6000 h and the results were compared to wrought databases. The data show that both creep strength and ductility were important factors in the overall creep performance of the sheets. Reductions in performance were observed due either to accelerated creep when grain size was fine or loss of rupture ductility when grain sizes approached the sheet thickness. Some combinations of heat-treatment and thickness were able to produce typical expected wrought creep properties. Historically a ‘rule-of-thumb’ requirement for creep testing suggests 3-5 grains per sample minimum dimension to ensure homogeneous behavior. This research shows that to ensure representative wrought creep performance (i.e. no effect of sample size), the sample minimum cross-section should be 10 times the average ASTM grain size. Statistical analysis of the microstructures suggests the population of larger grains as a controlling feature in creep failure.
The Behrens–Fisher problem occurs when testing the equality of means of two normal distributions without the assumption that the two variances are equal. This paper presents approaches based on the exact and near-exact distributions for the test statistic of the Behrens–Fisher problem, depending on different combinations of even or odd sample sizes. We present the exact distribution when both sample sizes are odd and the near-exact distribution when one or both sample sizes are even. The near-exact distributions are based on a finite mixture of generalized integer gamma (GIG) distributions, used as an approximation to the exact distribution, which consists of an infinite series. The proposed tests, based on the exact and the near-exact distributions, are compared with Welch’s t-test through Monte Carlo simulations, in particular for small and unbalanced sample sizes. The results show that the proposed approaches are competent solutions to the Behrens–Fisher problem, exhibiting precise sizes and better powers than Welch’s approach for those cases. Numerical studies show that the Welch’s t-test tends to be a bit more conservative than the test statistics based on the exact or near-exact distribution, in particular when sample sizes are small and unbalanced, situations in which the proposed exact or near-exact distributions obtain higher powers than Welch’s t-test.
The ASTM G72/G72M-15 Standard Test Method for Autogenous Ignition Temperature of Liquids and Solids in a High-Pressure Oxygen-Enriched Environment is currently used to evaluate materials for the ignition susceptibility driven by exposure to external heat in an enriched oxygen environment. Testing performed on highly volatile liquids such as cleaning solvents has proven problematic due to inconsistent test results (non-ignitions). Non-ignition results can be misinterpreted as favorable oxygen compatibility, although they are more likely associated with inadequate fuel-to-oxidizer ratios. Forced evaporation during purging and inadequate sample size were identified as two potential causes for inadequate available sample material during testing. In an effort to maintain adequate fuel-to-oxidizer ratios within the reaction vessel during test, several parameters were considered, including sample size, pretest sample chilling, pretest purging, and test pressure. Tests on a variety of solvents exhibiting a range of volatilities are presented in this paper. A proposed improvement to the standard test protocol as a result of this evaluation is also presented. Execution of the final proposed improved test protocol outlines an incremental step method of determining optimal conditions using increased sample sizes while considering test system safety limits. The proposed improved test method increases confidence in results obtained by utilizing the ASTM G72 autogenous ignition temperature test method and can aid in the oxygen compatibility assessment of highly volatile liquids and other conditions that may lead to false non-ignition results.
The ASTM G72/G72M-15 Standard Test Method for Autogenous Ignition Temperature of Liquids and Solids in a High-Pressure Oxygen-Enriched Environment is currently used to evaluate materials for the ignition susceptibility driven by exposure to external heat in an enriched oxygen environment. Testing performed on highly volatile liquids such as cleaning solvents has proven problematic due to inconsistent test results (non-ignitions). Non-ignition results can be misinterpreted as favorable oxygen compatibility, although they are more likely associated with inadequate fuel-to-oxidizer ratios. Forced evaporation during purging and inadequate sample size were identified as two potential causes for inadequate available sample material during testing. In an effort to maintain adequate fuel-to-oxidizer ratios within the reaction vessel during test, several parameters were considered, including sample size, pretest sample chilling, pretest purging, and test pressure. Tests on a variety of solvents exhibiting a range of volatilities are presented in this paper. A proposed improvement to the standard test protocol as a result of this evaluation is also presented. Execution of the final proposed improved test protocol outlines an incremental step method of determining optimal conditions using increased sample sizes while considering test system safety limits. The proposed improved test method increases confidence in results obtained by utilizing the ASTM G72 autogenous ignition temperature test method and can aid in the oxygen compatibility assessment of highly volatile liquids and other conditions that may lead to false non-ignition results.
This paper reports on an approach for minimizing data loads associated with satellite-acquired data, while improving the efficiency of global crop area estimates using remotely sensed, satellite-based data. Results of a sampling unit size investigation are given that include closed-form models for both nonsampling and sampling error variances. These models provide estimates of the sampling unit sizes that effect minimal costs. Earlier findings from foundational sampling unit size studies conducted by Mahalanobis, Jessen, Cochran, and others are utilized in modeling the sampling error variance as a function of sampling unit size. A conservative nonsampling error variance model is proposed that is realistic in the remote sensing environment where one is faced with numerous unknown nonsampling errors. This approach permits the sampling unit size selection in the global crop inventorying environment to be put on a more quantitative basis while conservatively guarding against expected component error variances.
In the repository are example scripts that perform uncertainty injection and propagation to flux balance analysis with outputs for a small sample size (for demonstration purpose only). For proper analysis, user should download the scripts and run for a large sample size (e.g., 10,000 samples). If you use the scripts, please cite the following Metabolic Engineering article: “Quantifying the propagation of parametric uncertainty on flux balance analysis” (https://doi.org/10.1016/j.ymben.2021.10.012) There are two subdirectories: /uncFBA/uncBiom: injection of normally distributed noise to biomass precursor coeffcients and ATP maintenance (growth-associated ATP maintenance (GAM) and non-growth associated ATP maintenance (NGAM)) /uncFBA/uncRHS: departure from steady-state by adding noise drawn from normal distribution to the RHS terms of mass balance constraints
A procedure to estimate wheat (Triticum aestivum L) area using sampling technique based on aerial photographs and digital LANDSAT MSS data is developed. Aerial photographs covering 720 square km are visually analyzed. To estimate wheat area, a regression approach is applied using different sample sizes and various sampling units. As the size of sampling unit decreased, the percentage of sampled area required to obtain similar estimation performance also decreased. The lowest percentage of the area sampled for wheat estimation with relatively high precision and accuracy through regression estimation is 13.90% using 10 square km as the sampling unit. Wheat area estimation using only aerial photographs is less precise and accurate than those obtained by regression estimation.
Laboratory-based Bidirectional Reflectance Distribution Function (BRDF) analysis of vegetation leaves, soil, and leaf litter samples is presented. The leaf litter and soil samples, numbered 1 and 2, were obtained from a site located in the savanna biome of South Africa (Skukuza: 25.0degS, 31.5degE). A third soil sample, number 3, was obtained from Etosha Pan, Namibia (19.20degS, 15.93degE, alt. 1100 m). In addition, BRDF of local fresh and dry leaves from tulip tree (Liriodendron tulipifera) and acacia tree (Acacia greggii) were studied. It is shown how the BRDF depends on the incident and scatter angles, sample size (i.e. crushed versus whole leaf,) soil samples fraction size, sample status (i.e. fresh versus dry leaves), vegetation species (poplar versus acacia), and vegetation s biochemical composition. As a demonstration of the application of the results of this study, airborne BRDF measurements acquired with NASA's Cloud Absorption Radiometer (CAR) over the same general site where the soil and leaf litter samples were obtained are compared to the laboratory results. Good agreement between laboratory and airborne measured BRDF is reported.
Computational models that formalize complex human behaviors enable study and understanding of such behaviors. However, collecting behavior data required to estimate the parameters of such models is often tedious and resource intensive. Thus, estimating dataset size as part of data collection planning (also known as Sample Size Determination) is important to reduce the time and effort of behavior data collection while maintaining an accurate estimate of model parameters. In this paper, we present a sample size determination method based on Uncertainty Quantification (UQ) for a specific Inverse Reinforcement Learning (IRL) model of human behavior, in two cases: 1) pre-hoc experiment design—conducted in the planning stage before any data is collected, to guide the estimation of how many samples to collect; and 2) post-hoc dataset analysis—performed after data is collected, to decide if the existing dataset has sufficient samples and whether more data is needed. Here, we validate our approach in experiments with a realistic model of behaviors of people with Multiple Sclerosis (MS) and illustrate how to pick a reasonable sample size target. Our work enables model designers to perform a deeper, principled investigation of effects of dataset size on IRL.
In upscaling methods, closures for nonlinear problems present a well-known challenge. While a number of theoretical methods have been proposed for handling such closures, nonlinearities still remain a significant obstacle for many problems. In this work, we use a combination of formal upscaling and data-driven machine learning for explicitly closing a nonlinear transport and reaction process in multiscale tissues. The classical effectiveness factor model is used to formulate the macroscale reaction kinetics. We train a multilayer perceptron network using training data generated by direct numerical simulations over microscale examples. Once trained, the network is used in an algorithm for numerically solving the upscaled (coarse-grained) differential equation describing mass transport and reaction in two example tissues. The network is described as being explicit in the sense that the network is trained using macroscale concentrations and gradients of concentration as components of the feature space rather than incorporating them as part of a constraint in the optimization process. Network training and solutions to the macroscale transport equations were computed for two different tissues. The two tissue types (brain and liver) exhibit markedly different geometrical complexity and spatial scale (cell size and sample size). The upscaled solutions for the average concentration are compared with numerical solutions derived from the microscale concentration fields by a posteriori averaging. There are three outcomes of this work of particular note. 1) Our overall approach results in an upscaled nonlinear PDE. The PDE is closed using a neural network, and our approach results in the definition of the classical effectiveness factor for effecting closure. 2) We identify particular source terms for the closure problem that are important for representing the structure of the closure. These source terms involve macroscale concentrations and their gradients. We adopt these source terms to use as explicit features in the learning algorithm. We find the trained networks that include the macroscale source terms generate models that are able to predict the correction factor with increased fidelity over those that do not. 3) We find that the trained network exhibits good generalizability, and it is able to predict the effectiveness factor with high fidelity for realistically-structured tissues despite the significantly different scale and geometrical complexity of the two example tissue types. This latter result emphasizes our purposeful connection between conventional averaging methods with the use of machine learning for closure; this contrasts with some machine learning methods for upscaling where the exact form of the macroscale equation remains unknown.