Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “generalized linear model”

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 595 records · Page 33
All sourcesNASA NTRSDOE OSTI
NASA NTRS1991

On the formulation of a minimal uncertainty model for robust control with structured uncertainty

In the design and analysis of robust control systems for uncertain plants, representing the system transfer matrix in the form of what has come to be termed an M-delta model has become widely accepted and applied in the robust control literature. The M represents a transfer function matrix M(s) of the nominal closed loop system, and the delta represents an uncertainty matrix acting on M(s). The nominal closed loop system M(s) results from closing the feedback control system, K(s), around a nominal plant interconnection structure P(s). The uncertainty can arise from various sources, such as structured uncertainty from parameter variations or multiple unsaturated uncertainties from unmodeled dynamics and other neglected phenomena. In general, delta is a block diagonal matrix, but for real parameter variations delta is a diagonal matrix of real elements. Conceptually, the M-delta structure can always be formed for any linear interconnection of inputs, outputs, transfer functions, parameter variations, and perturbations. However, very little of the currently available literature addresses computational methods for obtaining this structure, and none of this literature addresses a general methodology for obtaining a minimal M-delta model for a wide class of uncertainty, where the term minimal refers to the dimension of the delta matrix. Since having a minimally dimensioned delta matrix would improve the efficiency of structured singular value (or multivariable stability margin) computations, a method of obtaining a minimal M-delta would be useful. Hence, a method of obtaining the interconnection system P(s) is required. A generalized procedure for obtaining a minimal P-delta structure for systems with real parameter variations is presented. Using this model, the minimal M-delta model can then be easily obtained by closing the feedback loop. The procedure involves representing the system in a cascade-form state-space realization, determining the minimal uncertainty matrix, delta, and constructing the state-space representation of P(s). Three examples are presented to illustrate the procedure.

Belcastro, Christine M.↗
DOE OSTI2024

Data Imbalance, Uncertainty Quantification, and Transfer Learning in Data‐Driven Parameterizations: Lessons From the Emulation of Gravity Wave Momentum Transport in WACCM

Abstract Neural networks (NNs) are increasingly used for data‐driven subgrid‐scale parameterizations in weather and climate models. While NNs are powerful tools for learning complex non‐linear relationships from data, there are several challenges in using them for parameterizations. Three of these challenges are (a) data imbalance related to learning rare, often large‐amplitude, samples; (b) uncertainty quantification (UQ) of the predictions to provide an accuracy indicator; and (c) generalization to other climates, for example, those with different radiative forcings. Here, we examine the performance of methods for addressing these challenges using NN‐based emulators of the Whole Atmosphere Community Climate Model (WACCM) physics‐based gravity wave (GW) parameterizations as a test case. WACCM has complex, state‐of‐the‐art parameterizations for orography‐, convection‐, and front‐driven GWs. Convection‐ and orography‐driven GWs have significant data imbalance due to the absence of convection or orography in most grid points. We address data imbalance using resampling and/or weighted loss functions, enabling the successful emulation of parameterizations for all three sources. We demonstrate that three UQ methods (Bayesian NNs, variational auto‐encoders, and dropouts) provide ensemble spreads that correspond to accuracy during testing, offering criteria for identifying when an NN gives inaccurate predictions. Finally, we show that the accuracy of these NNs decreases for a warmer climate (4 × CO 2 ). However, their performance is significantly improved by applying transfer learning, for example, re‐training only one layer using ∼1% new data from the warmer climate. The findings of this study offer insights for developing reliable and generalizable data‐driven parameterizations for various processes, including (but not limited to) GWs.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1976

Thermal analyses of the International Ultraviolet Explorer (IUE) scientific instrument using the NASTRAN thermal analyzer (NTA): A general purpose summary

The NTA Level 15.5.2/3, was used to provide non-linear steady-state (NLSS) and non-linear transient (NLTR) thermal predictions for the International Ultraviolet Explorer (IUE) Scientific Instrument (SI). NASTRAN structural models were used as the basis for the thermal models, which were produced by a straight forward conversion procedure. The accuracy of this technique was sub-sequently demonstrated by a comparison of NTA predicts with the results of a thermal vacuum test of the IUE Engineering Test Unit (ETU). Completion of these tasks was aided by the use of NTA subroutines.

Jackson, C. E., Jr.↗
NASA NTRS2003

Transient Simulation of the Integrated Powerhead Demonstrator (IPD) Rocket Engine

The IPD rocket engine is the product of a joint Air Force/NASA program to demonstrate the concept of a full-flow staged combustion power cycle. In this type of rocket engine cycle, both the entire fuel flow and the entire oxidizer flow are combusted in one of two pre-bumers and used to drive two turbopumps, thus utilizing the flow work of the total propellant flow. The basis of the IPD engine program relies on integrating newly developed and pre-existing hardware to demonstrate the component and material technologies to make this concept feasible, while simultaneously saving development time and costs. To provide insight to the project team and contractors during engine development and test phases, the engine system was modeled at MSFC using the Rocket Engine Transient Simulation (ROCETS) software to analyze system performance and determine component integration issues. The ROCETS software is used extensively a MSFC to perform steady-state power-balances and transient simulations of thermodynamic power and general fluid systems. The software is favored for its capability to solve large systems of non-linear equations, its librarie of fluid properties and flow devices, its flexibility to modify existing code to improve the physics-derived approximations of real fluid thermodynamic behavior, and the ability to add unique system constraints. The purpose of this paper is to present the methodology used to model the IPD engine system, detail the pitfalls encountered with the software, and explain the approximations made to more accurately represent engine component and fluid behavior. Engine system performance output from the model will be presented and explained in comparison with real fluid and component behavior.

Guidos, Mike↗
NASA NTRS2020

The HadGEM3-GA7.1 radiative kernel: the importance of a well-resolved stratosphere

We present top-of-atmosphere and surface radiative kernels based on the atmospheric component (GA7.1) of the HadGEM3 general circulation model developed by the UK Met Office. We show that the utility of radiative kernels for forcing adjustments in idealised CO2 perturbation experiments is greatest where there is sufficiently high resolution in the stratosphere in both the target climate model and the radiative kernel. This is because stratospheric cooling to a CO2 perturbation continues to increase with height, and low-resolution or low-top kernels or climate model output are unable to fully resolve the full stratospheric temperature adjustment. In the sixth phase of the Coupled Model Intercomparison Project (CMIP6), standard atmospheric model data are available up to 1 hPa on 19 pressure levels, which is a substantial advantage compared to CMIP5. We show in the IPSL-CM6A-LR model where a full set of climate diagnostics are available that the HadGEM3-GA7.1 kernel exhibits linear behaviour and the residual error term is small, as well as from a survey of kernels available in the literature that in general low-top radiative kernels underestimate the stratospheric temperature response. The HadGEM3-GA7.1 radiative kernels are available at https://doi.org/10.5281/zenodo.3594673 (Smith, 2019).

radiative kernels↗
DOE OSTI2021

Cosmological constraints from H ii starburst galaxy, quasar angular size, and other measurements

ABSTRACT We compare the constraints from two (2019 and 2021) compilations of H ii starburst galaxy (H iiG) data and test the model independence of quasar (QSO) angular size data using six spatially flat and non-flat cosmological models. We find that the new 2021 compilation of H iiG data generally provides tighter constraints and prefers lower values of cosmological parameters than those from the 2019 H iiG data. QSO data by themselves give relatively model-independent constraints on the characteristic linear size, lm, of the QSOs within the sample. We also use Hubble parameter [H(z)], baryon acoustic oscillation (BAO), Pantheon Type Ia supernova (SN Ia) apparent magnitude (SN-Pantheon), and DES-3 yr binned SN Ia apparent magnitude (SN-DES) measurements to perform joint analyses with H iiG and QSO angular size data, since their constraints are not mutually inconsistent within the six cosmological models we study. A joint analysis of H(z), BAO, SN-Pantheon, SN-DES, QSO, and the newest compilation of H iiG data provides almost model-independent summary estimates of the Hubble constant, $H_0=69.7\pm 1.2\ \rm {km\,s^{-1}\,Mpc^{-1}}$, the non-relativistic matter density parameter, $\Omega _{\rm m_0}=0.293\pm 0.021$, and lm = 10.93 ± 0.25 pc.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Renormalization of Ising cage-net model and generalized foliation

A large class of type-I fracton models, including the X-cube model, have been found to be fixed points of the foliated renormalization group (RG). The system size of such foliated models can be changed by adding or removing decoupled layers of 2D topological states and continuous deformation of the Hamiltonian. In this paper, we study a closely related model—the Ising cage-net model—and find that this model is not foliated in the same sense. In fact, we point out certain unnatural restrictions in the foliated RG, and find that removing these restrictions leads to a generalized foliated RG under which the Ising cage-net model is a fixed point, and which includes the original foliated RG as a special case. The Ising cage-net model thus gives a prototypical example of the generalized foliated RG, and its system size can be changed either by condensing/uncondensing bosonic planon excitations near a 2D plane or through a linear-depth quantum circuit in the same plane. Additionally, we show that these two apparently different RG procedures are closely related, as they lead to the same gapped boundary when implemented in part of a plane. Finally, we briefly discuss the implications for foliated fracton phases, whose universal properties will need to be reexamined in light of the generalized foliated RG.

2-dimensional systems↗
DOE OSTI2021

Whetstone v.0.9.2

Whetstone is a deep learning software library for spiking/binary threshold neuromorphic hardware. Built to integrate with Keras models, Whetstone provides a collection of layers, callbacks, and utility functions designed to allow the training of integrate-and-fire neurons rather than traditional analog activation functions (e.g. rectified linear units). Generally, training a network for neuromorphic hardware is challenging due to the discontinuity of the activation function. Whetstone overcomes this challenge by using ever-closer continuous approximations of a threshold activation, modified during training time. This allows rapid and consistent training for ultra-low powered neuromorphic hardware.

Severa, William↗
DOE OSTI2025

Nuclear Data Adjustment for Nonlinear Applications in the OECD/NEA WPNCS SG14 Benchmark -- A Bayesian Inverse UQ-based Approach for Data Assimilation

The Organization for Economic Cooperation and Development (OECD) Working Party on Nuclear Criticality Safety (WPNCS) proposed a benchmark exercise to assess the performance of current nuclear data adjustment techniques applied to nonlinear applications and experiments with low correlation to applications. This work introduces Bayesian Inverse Uncertainty Quantification (IUQ) as a method for nuclear data adjustments in this benchmark, and compares IUQ to the more traditional methods of Generalized Linear Least Squares (GLLS) and Monte Carlo Bayes (MOCABA). Posterior predictions from IUQ showed agreement with GLLS and MOCABA for linear applications. When comparing GLLS, MOCABA, and IUQ posterior predictions to computed model responses using adjusted parameters, we observe that GLLS predictions fail to replicate computed response distributions for nonlinear applications, while MOCABA shows near agreement, and IUQ uses computed model responses directly. We also discuss observations on why experiments with low correlation to applications can be informative to nuclear data adjustments and identify some properties useful in selecting experiments for inclusion in nuclear data adjustment. Performance in this benchmark indicates potential for Bayesian IUQ in nuclear data adjustments.

FOS: Computer and information sciences↗
NASA NTRS1976

Aspects of job scheduling

A mathematical model for job scheduling in a specified context is presented. The model uses both linear programming and combinatorial methods. While designed with a view toward optimization of scheduling of facility and plant operations at the Deep Space Communications Complex, the context is sufficiently general to be widely applicable. The general scheduling problem including options for scheduling objectives is discussed and fundamental parameters identified. Mathematical algorithms for partitioning problems germane to scheduling are presented.

Phillips, K.↗
NASA NTRS1980

Linear continuous and sampled-data systems

Program performs general analysis of linear and continuous, discrete and sampled-data systems using state-variable techniques. Program is especially suited analysis of linearized control system problems. It also can be used to model system described by combination of differential equations and Laplace transform blocks, such as aircraft control system.

Edwards, J. W.↗
NASA NTRS1978

Dynamic model verification of large structural systems

The objective of the present methodology is two-fold: (1) to process test data obtained from either modal survey tests, or slow sine-sweep tests, to extract a set of orthogonal modes best matching the test data while being commensurate with the dynamic model, and (2) to modify submatrices of the dynamic model mass and stiffness matrices to adjust the model to best fit the test data. The method has been implemented using a linear statistical sequential estimator for computation on a CDC computer. Demonstration problems involving Space Shuttle quarter-scale vibration test data and dynamic models have been run. This paper will discuss the general methodology and experience to date.

Lee, L. T.↗
NASA NTRS1985

Improving the absolute positioning accuracy of robot manipulators

The absolute positioning accuracy of robot manipulator can be increased substantially by updating the nominal link parameters in the control software. This paper presents a general method to estimate the link parameter errors for any serial link manipulator (i.e., n links, and any combination of revolute and prismatic joints). The parameters are estimated through a linear kinematic model which relates the link bias errors to the end-effector positioning error. Only end-effector measurements are required instead of individual link measurements to implement this method.

Hayati, S.↗
NASA NTRS1989

CLAES focal plane array

The Cryogenic Limb Array Etalon Spectrometer for the NASA Upper Atmospheric Research Satellite uses solid-state focal plane arrays to detect emission from the earth's atmosphere over the IR wavelength range 3.5 to 13 microns. This paper discusses the design of the focal plane detector assembly and compares calculated performance with measurements. Measurements were made of focal plane noise and responsivity as functions of frequency (2 to 500 Hz) and temperature (12 to 19 K), pixel-to-pixel and across-array crosstalk, and linearity over a dynamic range of 100,000. The measurements demonstrate that the arrays satisfy the science requirements, and that, in general, there is reasonable agreement between the measurements and the analytical model.

Roche, A. E.↗
NASA NTRS1995

Cosmic ray LET spectra and doses on board Cosmos-2044 biosatellite

Results of the experiments on board Cosmos-2044 (Biosatellite 9) are presented. Various nuclear track detectors (NTD) (dielectric, AgCl-based, nuclear emulsions) were used to obtain the Linear Energy Transfer (LET) spectra inside and outside the satellite. The spectra from the different NTDs have proved to be in general agreement. The results of LET spectra calculations using two different models are also presented. The resultant LET distributions are used to calculate the absorbed and equivalent doses and the orbit-averaged quality factors (QF) of the cosmic rays (CR). Absorbed dose rates inside (approximately 20 g cm (exp -2) shielding) and outside (1 g cm(exp -2) the spacecraft, omitting electrons, were found to be 4.8 and 8.6 mrad d (exp -1), respectively, while the corresponding equivalent doses were 8.8 and 19.7 mrem d(exp -1). The effects of the flight parameters on the total fluence of, and on the dose from the CR particles are analyzed. Integral dose distributions of the detected particles are also determined. The LET values which separate absorbed and equivalent doses into 50% intervals are estimated. The CR-39 dielectric NTD is shown to detect 20-30% of the absorbed dose and 60-70% of the equivalent dose in the Cosmos-2044 orbit. The influence of solar activity phase on the magnitude of CR flux is discussed.

Watts, J. W., Jr.↗
NASA NTRS1995

Multiparameter linear least-squares fitting to Poisson data one count at a time

A standard problem in gamma-ray astronomy data analysis is the decomposition of a set of observed counts, described by Poisson statistics, according to a given multicomponent linear model, with underlying physical count rates or fluxes which are to be estimated from the data. Despite its conceptual simplicity, the linear least-squares (LLSQ) method for solving this problem has generally been limited to situations in which the number n(sub i) of counts in each bin i is not too small, conventionally more than 5-30. It seems to be widely believed that the failure of the LLSQ method for small counts is due to the failure of the Poisson distribution to be even approximately normal for small numbers. The cause is more accurately the strong anticorrelation between the data and the wieghts w(sub i) in the weighted LLSQ method when square root of n(sub i) instead of square root of bar-n(sub i) is used to approximate the uncertainties, sigma(sub i), in the data, where bar-n(sub i) = E(n(sub i)), the expected value of N(sub i). We show in an appendix that, avoiding this approximation, the correct equations for the Poisson LLSQ (PLLSQ) problems are actually identical to those for the maximum likelihood estimate using the exact Poisson distribution. We apply the method to solve a problem in high-resolution gamma-ray spectroscopy for the JPL High-Resolution Gamma-Ray Spectrometer flown on HEAO 3. Systematic error in subtracting the strong, highly variable background encountered in the low-energy gamma-ray region can be significantly reduced by closely pairing source and background data in short segments. Significant results can be built up by weighted averaging of the net fluxes obtained from the subtraction of many individual source/background pairs. Extension of the approach to complex situations, with multiple cosmic sources and realistic background parameterizations, requires a means of efficiently fitting to data from single scans in the narrow (approximately = 1.2 keV, HEAO 3) energy channels of a Ge spectrometer, where the expected number of counts obtained per scan may be very low. Such an analysis system is discussed and compared to the method previously used.

Wheaton, Wm. A.↗
NASA NTRS2017

An Investigation of Dust Storms Observed with the Mars Color Imager

Daily global imaging by the Mars Color Imager (MARCI) continues the record of the Mars Orbiter Camera (MOC) and has allowed creation of a long-duration record of Martian dust storms. We observe dust storms over the first two Mars years of the MARCI record, including tracking individual storms over multiple sols, as well as tracking the growth and recession of the seasonal polar caps. Using the combined 6 Mars year record of textured dust storms (storms with visible textures on the observed dust cloud tops), we study the relationship between textured dust storm activity and meteorology (as simulated by the MarsWRF general circulation model) and surface properties. We find that textured dust storms preferentially occur in places and seasons with above average surface wind stress. Textured dust storm occurrence also has a modest linear anti-correlation with surface albedo (0.43) and topography (0.40). Lastly, we perform an empirical orthogonal function (EOF) analysis on the distribution of occurrence of textured dust storms and find that over 50 of the variance in textured dust storm activity can be explained by two EOF modes. We associate the first EOF mode with cap-edge storms just before Ls = 180deg and the second EOF mode with flushing dust storms that occur from Ls = 180-210deg and again near Ls = 320deg.

Guzewich, Scott D.↗
NASA NTRS2021

Transition Analysis for the CRM-NLF Wind Tunnel Configuration using Transport Equation Models and Linear Stability Correlations

Transition models based on auxiliary transport equations augmenting the Reynolds-averaged Navier-Stokes (RANS) framework rely upon transition correlations that were derived from a limited number of low-speed experiments. Furthermore, these models often account for only a subset of the relevant transition mechanisms and/or cannot accurately predict the sensitivity of those mechanisms to the changes in significant flow parameters. A preceding investigation had targeted the assessment of the transport-equation-based transition models in NASA's OVERFLOW 2.3b solver, namely, the amplification factor transport (AFT-2017b) equation model coupled with the Spalart-Allmaras RANS model and the Langtry-Menter transition models (LM2009 without crossflow effects and LM2015 including the modeling of crossflow transition) implemented with Menter’s shear-stress transport equation (SST2003) RANS model. Comparisons with recent measurements at transonic freestream conditions on the Common Research Model with Natural Laminar Flow (CRM-NLF) reinforced our earlier finding that all three of the above models significantly underpredict the reported extent of the laminar flow region over the entire span of the wing, regardless of the dominant instability mechanism(s) underlying the onset of the transition process. The underprediction of the laminar flow extent was attributed to the failure of the above models in accounting for the stabilizing effect of compressibility on the amplification of Tollmien-Schlichting instabilities. Based on previous linear stability studies related to compressibility effects, the present work proposes modifications to the two classes of transition models that reduce to the original form of each model at low subsonic speeds and do not require any nonlocal flow information or additional transport equation(s). The modifications are shown to significantly improve the predicted laminar extent of the flow and compare well against the data from the CRM-NLF experiment. Additionally, a previous assessment of transition prediction based on the dual, nonparallel N -factor method in conjunction with linear parabolized stability equations (PSE) is extended to additional angles of attack to provide the first comprehensive assessment of transition models based on nonparallel disturbance amplification over the CRM-NLF. In general, the transition criterion based on the dual, nonparallel N-factor method with N TS = N CF = 6 is reasonably successful at correlating with the measured transition fronts at R eMAC = 15 million for all angles of attack investigated herein and provides additional validation of the improved predictions from the compressibility-corrected transition models.

CFD modeling↗