Search NASA⌕ Search

SEARCH · Search NASA

Results for “regression to the mean”

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 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2020

The Influence of Frequency and Duration Of Seismic Ground Motion On The Size Of Triggered Landslides—A Regional View

Observation, theory, and intuition all suggest that larger earthquakes should trigger larger landslides. Many factors could contribute to this, including depth-dependent shear strength or non-linearity of ground motion in soils and rock, but we hypothesize that the key characteristics of large earthquakes causing this phenomenon are (in addition to magnitude) the frequency and duration of the strong ground motion. Because of the paucity of site-specific data for detailed analysis, we take a regional approach to this question by analyzing strong-motion records and earthquake-induced landslide (EQIL) inventories from six well-documented earthquakes. Ground motion is characterized using earthquake magnitude and the median durations and frequencies (mean periods) of subsets of strong-motion records relevant to landslide triggering. EQIL inventories are characterized using the median landslide area of the entire inventory as well as the median areas of the largest 1% of the landslides and the largest 10 landslides. We then compare ground-motion characteristics with landslide size statistics to determine possible correlations. Comparisons of all earthquake- and landslide-size statistics show strong positive correlations between landslide size and (1) magnitude, (2) ground-motion duration, and (3) mean period. Although all the ground-motion measures yield highly correlated regressions, mean period appears to be the best overall predictor of landslide size. Landslide modeling using Newmark's sliding-block method also shows that longer mean periods and durations and larger magnitudes correlate strongly with increases in modeled displacements. These results support our hypothesis that increasing period and duration of seismic ground motion are the physical factors driving increased landslide sizes for larger earthquakes. Additional studies including data from a much larger set of earthquakes is needed to confirm the results of this initial study.

Hakan Tanyas↗
NASA NTRS1993

Cerebral blood flow velocity in humans exposed to 24 h of head-down tilt

This study investigates cerebral blood flow (CBF) velocity in humans before, during, and after 24 h of 6 deg head-down tilt (HDT), which is a currently accepted experimental model to simulate microgravity. CBF velocity was measured by use of the transcranial Doppler technique in the right middle cerebral artery of eight healthy male subjects. Mean CBF velocity increased from the pre-HDT upright seated baseline value of 55.5 +/- 3.7 (SE) cm/s to 61.5 +/- 3.3 cm/s at 0.5 h of HDT, reached a peak value of 63.2 +/- 4.1 cm/s at 3 h of HDT, and remained significantly above the pre-HDT baseline for over 6 h of HDT. During upright seated recovery, mean CBF velocity decreased to 87 percent of the pre-HDT baseline value. Mean CBF velocity correlated well with calculated intracranial arterial pressure (IAP). As analyzed by linear regression, mean CBF velocity = 29.6 + 0.32IAP. These results suggest that HDT increases CBF velocity by increasing IAP during several hours after the onset of microgravity. Importantly, the decrease in CBF velocity after HDT may be responsible, in part, for the increased risk of syncope observed in subjects after prolonged bed rest and also in astronauts returning to Earth.

Kawai, Y.↗
NASA NTRS1984

Use of Thematic Mapper for water quality assessment

The evaluation of simulated TM data obtained on an ER-2 aircraft at twenty-five predesignated sample sites for mapping water quality factors such as conductivity, pH, suspended solids, turbidity, temperature, and depth, is discussed. Using a multiple regression for the seven TM bands, an equation is developed for the suspended solids. TM bands 1, 2, 3, 4, and 6 are used with logarithm conductivity in a multiple regression. The assessment of regression equations for a high coefficient of determination (R-squared) and statistical significance is considered. Confidence intervals about the mean regression point are calculated in order to assess the robustness of the regressions used for mapping conductivity, turbidity, and suspended solids, and by regressing random subsamples of sites and comparing the resultant range of R-squared, cross validation is conducted.

Horn, E. M.↗
NASA NTRS2013

Modeling Longitudinal Data Containing Non-Normal Within Subject Errors

The mission of the National Aeronautics and Space Administration’s (NASA) human research program is to advance safe human spaceflight. This involves conducting experiments, collecting data, and analyzing data. The data are longitudinal and result from a relatively few number of subjects; typically 10 – 20. A longitudinal study refers to an investigation where participant outcomes and possibly treatments are collected at multiple follow-up times. Standard statistical designs such as mean regression with random effects and mixed–effects regression are inadequate for such data because the population is typically not approximately normally distributed. Hence, more advanced data analysis methods are necessary. This research focuses on four such methods for longitudinal data analysis: the recently proposed linear quantile mixed models (lqmm) by Geraci and Bottai (2013), quantile regression, multilevel mixed–effects linear regression, and robust regression. This research also provides computational algorithms for longitudinal data that scientists can directly use for human spaceflight and other longitudinal data applications, then presents statistical evidence that verifies which method is best for specific situations. This advances the study of longitudinal data in a broad range of applications including applications in the sciences, technology, engineering and mathematics fields.

Feiveson, Alan↗
DOE OSTI2023

Presentation: Sensor Anomaly Detection for Nuclear Reactor Systems Utilizing Linear Regression and K-Means Unsupervised Machine Learning: An overview of methods and results

This presentation is a culmination of work which has occurred over the course of a 10-week internship. Anomaly detection methods must be both robust enough to detect subtle anomalies yet not so sensitive to report false positives, which would result significant loss of revenue. Methods currently being developed for autonomous systems are often pursuing a Digital Twin method, which will look at the entire system and model it as a whole. This presentation, however, focuses less on direct application to an NPP, rather acting as a proof of concept for the methods developed. For the project, we look to develop methods to analyze steady-state data and report anomalies.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2023

Sensor Anomaly Detection for Nuclear Reactor Systems Utilizing Linear Regression and K-Means Unsupervised Machine Learning

Nuclear reactors and related systems are becoming increasingly complex due to advancing technologies in next-generation power reactors. This increased complexity necessitates enhanced automation and data management capabilities. To successfully realize autonomous systems, methods must be developed to handle vast volumes of data and effectively distinguish anomalous data from noise and expected data. While impressive models utilizing digital twins and similar approaches are under development, here we propose a simplified model for analyzing fundamental methods and techniques. Initially, we created a general dataset by using initial data from PCTRAN in order to represent ideal steady-state conditions. We then inserted anomalies based on prevalent sensor anomaly types (e.g., point anomalies, linear drift, and downward deviations), along with unusual anomalies such as exponential drift and upward deviations. To detect anomalies, we developed a program that employs data partitioning and linear regression to preprocess and filter the anomalous data. A K-Means machine learning (ML) method was then applied to separate and count the data within the anomalous partition. The results from all datasets—apart from exponential growth—demonstrated positive outcomes, with each returning multiple instances of greaterthan-95% accuracy. We conducted further investigations using Idaho National Laboratory’s RAVEN software to perform a sensitivity analysis on the input variables (R 2 Tolerance, Slope Tolerance, and Window Size) and found that the output variables (Accuracy and Time) were most sensitive to the Window Size. Despite the promising results published, further development is required to effectively apply these methods to nuclear systems. Nevertheless, the strengths of this approach are evident and hold promise for future applications in the field.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
NASA NTRSDate not supplied

Monthly Mean DNI and GTI Derived from Monthly Mean GHI and DHI Using Two Methods: Comparisons with the BSRN Data and the Results Derived from the CERES Hourly Data

Monthly mean Global Horizontal Irradiances (GHI) and Diffuse Horizontal Irradiances (DHI) are more widely available than monthly mean Direct Normal Irradiances (DNI) and Global Tilted Irradiances (GTI). Empirical methods have been developed to derive monthly mean DNI and GTI from monthly mean GHI or from GHI and DHI. In this paper, we evaluate two such methods. The first one was the Whitlock Method developed by Charles H. Whitlock (2005) for the NASA POWER project by means of regression of the BSRN data. The method expresses the monthly mean DHI-to-GHI ratio as polynomial functions of monthly mean clearness index, sunset hour angle and noon solar elevation angle on the monthly-average-day. The monthly mean DNI is calculated by dividing the monthly mean GHI-DHI difference, or DirHI, by the cosine of the solar zenith angle at the mid-time between sunrise and solar noon on the monthly-average-day. The second method is the LJCR Method developed by Liu and Jorden (1960) and Collares-Pereira and Rabl (1979), and this method empirically splits monthly mean GHI and DHI into hourly means on the monthly-average-day, and the resulting hourly mean GHI and DHI and their difference, DirHI, can then be used to compute the monthly mean DNI, GTI and the global solar tracker irradiance (GTrI). This method is also used by RETScreen. We recently produced a set of hourly DNI and DHI by bias-correcting the CERES hourly DNI and DHI, and computed hourly GTI and GTrI as well. The data span twenty plus years from March 2000 to near present on a 1 by 1 grid system. The monthly mean CERES GHI and the corrected DHI are used as inputs to the above two methods to compute monthly mean DNI, GTI and GTrI. Through comparisons with the BSRN data, it is found that the Whitlock Method, with slight modification, and the LJCR Method can produce results that are nearly as good as the results derived from the CERES hourly data.

Taiping Zhang↗
NASA NTRS1995

CD4(+)/CD8(+) T-lymphocyte Ratio: Effects of Rehydration before Exercise in Dehydrated Men

Effects of fluid ingestion on CD4+/CD8+ T-lymphocyte cell ratios were measured in four dehydrated men (ages 30-46 yr) before and after 70 min of supine submaximal (71 % VO(sub 2max) lower extremity cycle exercise. Just before exercise, Evans blue dye was injected for measurement of plasma volume. The subjects then drank one of six fluid formulations (12 ml/kg) in 3-4 min. All six mean post-hydration (pre-exercise) CD4+/CD8+ ratios (Becton-Dickinson Fluorescence Activated Cell Sorter and FACScan Consort-30 software program were below the normal range of 1.2-1.5; mean (+/- SE) and range were 0.77 +/- 0.12 and 0.39-1.15, respectively. The post-exercise ratios increased: mean = 1.36 =/- 0.15 (P less than 0.05) and range = 0.98-1.98. Regression of mean CD4+/CD8+ ratios on mean plasma osmolality resulted in pre- and post-exercise correlation coefficients of -0.76 (P less than 0.10) and -0.92 (P less than 0.01), respectively. The decreased pre-exercise ratios (after drinking) were probably not caused by the Evans blue dye but appeared to be associated more with the stress (osmotic) of dehydration. The increased post-exercise ratios to normal levels accompanied the rehydration and were not due to the varied electrolyte and osmotic concentrations of the ingested fluids or to the varied vascular volume shifts during exercise. Thus, the level of subject hydration and plasma osmotality may be factors involved in the mechanism of immune system modulation induced by exercise.

Greenleaf, John E.↗
NASA NTRS1992

Hybrid rocket fuel combustion and regression rate study

The objectives of this study are to develop hybrid fuels (1) with higher regression rates and reduced dependence on fuel grain geometry and (2) that maximize potential specific impulse using low-cost materials. A hybrid slab window motor system was developed to screen candidate fuels - their combustion behavior and regression rate. Combustion behavior diagnostics consisted of video and high speed motion pictures coverage. The mean fuel regression rates were determined by before and after measurements of the fuel slabs. The fuel for this initial investigation consisted of hydroxyl-terminated polybutadiene binder with coal and aluminum fillers. At low oxidizer flux levels (and corresponding fuel regression rates) the filled-binder fuels burn in a layered fashion, forming an aluminum containing binder/coal surface melt that, in turn, forms into filigrees or flakes that are stripped off by the crossflow. This melt process appears to diminish with increasing oxidizer flux level. Heat transfer by radiation is a significant contributor, producing the desired increase in magnitude and reduction in flow dependency (power law exponent) of the fuel regression rate.

Strand, L. D.↗
NASA NTRS1997

Cycle-Powered Short Radius (1.9 m) Centrifuge: Effect of Exercise Versus Passive Acceleration on Heart Rate in Humans

In addition to extensive use of lower extremity physical exercise training as a countermeasure for the work capacity component of spaceflight deconditioning, some form of additional head-to-foot (+Gz) gravitational (orthostatic) stress may be required to further attenuate or prevent the signs and symptoms (nausea, vertigo, instability, fatigue) of the general reentry syndrome (GRS) that can reduce astronaut performance during landing. Orthostatic (head-to-foot) stress can be induced by standing, by lower body negative pressure, and by +Gz acceleration. One important question is whether acceleration training alone or with concurrent leg exercise would provide sufficient additive stimulation to attenuate the GRS. Use of a new human-powered centrifuge may be the answer. Thus, the purpose for this study was to compare heart rate (HR), i.e., a stress response during human-powered acceleration, in four men (35-62 yr) and two women (30-31 yr) during exercise acceleration versus passive acceleration (by an off-board operator) at 100% (maximal acceleration = A(max)), and at 25%, 50%, and 75% of A(max). Mean (+/-SE) A(max) was 43.7 +/- 1.3 rpm (+3.9 +/- 0.2Gz). Mean HR at exercise A(max) was 189 +/- 13 b/min (50-70 sec run time), and 142 +/- 22 b/min at passive A(max) (40-70 sec run time). Regression of mean HR on the various +Gz levels indicated explained variance (correlations squared) of r(exp 2) = 0.88 (exercise) and r(exp 2) = 0.96 (passive): exercise HR of 107 +/- 4 (25%) to 189 +/- 13 (100%) b/min were 43-50 b/min higher (p less than 0.05) than comparable passive HR of 64 +/- 2 to 142 +/- 22 b/min. Thus, exercise adds significant physiological stress during +Gz acceleration. Inflight use of this combined exercise and acceleration countermeasure may maintain work capacity as well as normalize acceleration and orthostatic tolerances which could attenuate or perhaps eliminate the GRS.

Greenleaf, J. E.↗
DOE OSTI2024

Learning Functions Varying along a Central Subspace

Many functions of interest are in a high-dimensional space but exhibit low-dimensional structures. This paper studies regression of an s-Hölder function in $R^D$ which varies along a central subspace of dimension $d$ while $d \ll D$. A direct approximation of $f$ in $R^D$ with an accuracy $\varepsilon$ requires the number of samples in the order of $\varepsilon^{-(2s+D)/s}$. In this paper, we analyze the generalized contour regression (GCR) algorithm for the estimation of the central subspace and use piecewise polynomials for function approximation. GCR is among the best estimators for the central subspace, but its sample complexity is an open question. In this paper, we partially answer this questions by proving that if a variance quantity is exactly known, GCR leads to a mean squared estimation error of $O(n^{-1})$ for the central subspace. The estimation error of this variance quantity is also given in this paper. The mean squared regression error of $f$ is proved to be in the order of $(n/\log n)^{-\frac{2s}{2s+d}}$, where the exponent depends on the dimension of the central subspace instead of the ambient space . This result demonstrates that GCR is effective in learning the low-dimensional central subspace. We also propose a modified GCR with improved efficiency. Here, the convergence rate is validated through several numerical experiments.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Electrical Conductivities and Association Constants in Dilute Aqueous NdCl 3 Solutions from 298 to 523 K along an Isobar of 25 MPa

Here, electrical measurements were performed in dilute aqueous NdCl 3 solutions from 298 to 523 K along the 25 MPa isobar to obtain limiting conductivities and association constants. The specific conductivity data were estimated using a continuous flow cell and a Markov Chain Monte Carlo (MCMC) correction algorithm. The limiting conductivities for the salts in water were derived by regressing the mean spherical approximation model and speciation analyses based on the MCMC algorithm and the deep earth water model. The limiting conductivities derived from the experimental data were correlated using the Zimmerman-Arcis-Tremaine correlation and agreed well with a predictive correlation proposed by Smolyakov, Anderko, and Lencka. Only the first association constant between neodymium and chloride could be derived at low temperatures (<373 K) due to the apparent large statistical uncertainty of the second association constant. Above 373 K, both association constants could be derived and show a reasonable agreement with Migdisov and Williams-Jones and Gammons et al.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

An finite element analysis surrogate model with boundary oriented graph embedding approach for rapid design

Abstract In this work, we present a boundary oriented graph embedding (BOGE) approach for the graph neural network to assist in rapid design and digital prototyping. The cantilever beam problem has been solved as an example to validate its potential of providing physical field results and optimized designs using only 10 ms. Providing shortcuts for both boundary elements and local neighbor elements, the BOGE approach can embed unstructured mesh elements into the graph and performs an efficient regression on large-scale triangular-mesh-based finite element analysis (FEA) results, which cannot be realized by other machine-learning-based surrogate methods. It has the potential to serve as a surrogate model for other boundary value problems. Focusing on the cantilever beam problem, the BOGE approach with 3-layer DeepGCN model achieves the regression with mean square error (MSE) of 0.011 706 (2.41% mean absolute percentage error) for stress field prediction and 0.002 735 MSE (with 1.58% elements having error larger than 0.01) for topological optimization. The overall concept of the BOGE approach paves the way for a general and efficient deep-learning-based FEA simulator that will benefit both industry and Computer Aided Design (CAD) design-related areas.

42 ENGINEERING↗
NASA NTRS1976

Wind tunnel test on a 1/4.622 Froude scale, hingeless rotor, tilt rotor model, volume 1

Wing tunnel test data on a 1/4.622 Froude scale, hingeless rotor, tilt rotor mode are reported for all potential flight conditions through hover and a wide envelope of transitions. A mathematical model was used to describe the rotor system in real time simulation by means of regression analyses. Details of the model, test program and data system are provided together with four data files for hover and transition.

Magee, J. P.↗
NASA NTRS1983

Impact of LANDSAT MSS sensor differences on change detection analysis

Some 512 by 512 pixel subwindows for simultaneously acquired scene pairs obtained by LANDSAT 2,3 and 4 multispectral band scanners were coregistered using LANDSAT 4 scenes as the base to which the other images were registered. Scattergrams between the coregistered scenes (a form of contingency analysis) were used to radiometrically compare data from the various sensors. Mode values were derived and used to visually fit a linear regression. Root mean square errors of the registration varied between .1 and 1.5 pixels. There appear to be no major problem preventing the use of LANDSAT 4 MSS with previous MSS sensors for change detection, provided the noise interference can be removed or minimized. Data normalizations for change detection should be based on the data rather than solely on calibration information. This allows simultaneous normalization of the atmosphere as well as the radiometry.

Likens, W. C.↗
NASA NTRS1972

Operational Weight Estimations of Commercial Jet Transport Aircraft

In evaluating current or proposed commercial transport airplanes, there has not been available a ready means to determine weights so as to compare airplanes within this particular class. This paper describes the development of and presents such comparative tools. The major design characteristics of current American jet transport airplanes were collected, and these data were correlated by means of regression analysis to develop weight relationships for these airplanes as functions of their operational requirements. The characteristics for 23 airplanes were assembled and examined in terms of the effects of the number of people carried, the cargo load, and the operating range. These airplane characteristics were correlated for the airplanes as one of three subclasses, namely the small, twin-engine jet transport, the conventional three- and four-engine jets, and the new wide-body jets.

Anderson, Joseph L.↗
NASA NTRS1993

Horizontal angular VOR, nystagmus dumping, and sensation duration in spacelab SLS-1 crewmembers

In 1G, the apparent time constant (Td) of postrotatory SPV decay with the head tilted face down is 55% of that with head erect (Te). This phenomenon is called "nystagmus dumping" and has been attributed to G effects on VOR velocity storage. Similarly, postrotatory sensation duration with head tilted (Dd) is 32% of that when head erect (De). In parabolic flight, Te and De are 70% of 1-G values, but a pitch back dumping movement produces no further change. Te, Td, and Dd have not previously been measured in orbital flight. VOR and sensation duration was tested in 4 crewmembers in 4 preflight, 1 inflight (days 4 or 5) and 4 post flight sessions. Bitemporal EOG was recorded with eyes open in darkness. Instructions were to "gaze straight ahead," and indicate when "rotation sensation disappears or becomes ambiguous". Subjects were rotated CW and CCW head erect for 1 min at 120 degrees/s, stopped, and EOG was recorded for another 1 min. This procedure was then used to study dumping, except that immediately after chair stop, subjects pitched their head forward 90 degrees. SPV was calculated using order statistic filtering, and dropouts removed using an iterative model fitting method. Te and Td were determined by logarithmic linear regression of mean SPV for each subject. In orbit, 90 degrees pitch movement produced rapid subjective dumping, but not nystagmus dumping. Dd was noticeably shorter ("almost instantaneous") compared to preflight Dd. Te and Td in orbit were similar to preflight Te for 3/4 subjects (rather than to preflight Td as expected). No consistent VOR gain changes were seen in orbit. Although Te is known to decrease acutely in parabolic flight, a longer time constant was measured in 3/4 subjects after 4-5 days adaptation to weightlessness, suggesting a return of angular velocity storage.

NASA Experiment Number 178072 1/2↗
NASA NTRS2019

Lightning NOx Production in the Tropics as Determined Using OMI NO2 Retrievals and WWLLN Stroke Data

Nitrogen oxide (NOx) production by lightning in the tropics is estimated using tropospheric NOx amounts (LNOx*) over deep convective grid boxes derived from OMI (Ozone Monitoring Instrument) nitrogen dioxide (NO2) slant columns and detection efficiency adjusted WWLLN (World Wide Lightning Location Network) flashes. The lightning NOx production efficiency (LNOx PE) in the tropics is determined for the austral and boreal summers of 2007 to 2011 by regressing regional mean daily values of LNOx* for individual seasons against daily flash totals during flash windows prior to the OMI overpass. LNOx PE is determined to be approximately two times larger over marine locations than over continental locations possibly because marine flashes are more energetic. Overall, the mean LNOx PE for the tropics is calculated to be 170 ± 100 mol per flash with values over the tropical Pacific (low flash rate region) being largest. The main contributors to uncertainties in PE are uncertainties in WWLLN flash detection efficiency, upper tropospheric NOx lifetime in the near field of convection, and air mass factor biases.

Dale J. Allen↗