SEARCH · Search NASA

Results for “data reduction algorithms”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

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

At least 253 records · Page 14

All sources NASA NTRS DOE OSTI

NASA NTRS1989

Compression Of Data In Imaging Radar Polarimetry

Algorithms developed to reduce number of radar polarimetric data processed to synthesize image of arbitrary combination of transmitting and receiving polarizations. Brings image-processing requirements within computing capabilities of typical users, without degrading images excessively. In scattering-matrix approach to reduction of image data, four adjacent picture elements combined into one by synthesizing new scattering matrix from scattering matrices of four elements. In phase-matrix approach, phase matrices generated from scattering matrices of four adjacent picture elements, and four phase matrices added to combine four picture elements into one.

Zebker, H. A.↗

NASA NTRS2014

JPSS Science Data Services for the Direct Readout Community

The Suomi National Polar-orbiting Partnership (S-NPP) and Joint Polar Satellite System (JPSS) High Rate Data (HRD) link provides Direct Broadcast data to users in real-time, utilizing their own remote field terminals. The Field Terminal Support (FTS) provides the resources needed to support the Direct Readout communities by providing software, documentation, and periodic updates to enable them to produce data products from SNPP and JPSS. The FTS distribution server will also provide the necessary ancillary and auxiliary data needed for processing the broadcasts, as well as making orbital data available to assist in locating the satellites of interest. In addition, the FTS provides development support for the algorithm and software through GSFC Direct Readout Laboratory (DRL) International Polar Orbiter Processing Package (IPOPP) and University of Wisconsin (UWISC) Community Satellite Processing Package (CSPP), to enable users to integrate the algorithms into their remote terminals. The support the JPSS Program provides to the institutions developing and maintaining these two software packages, will demonstrate the ability to produce ready-to-use products from the HRD link and provide risk reduction effort at a minimal cost. This paper discusses the key functions and system architecture of FTS.

Data Processing↗

NASA NTRS1984

An algorithm for maximum likelihood estimation using an efficient method for approximating sensitivities

An algorithm for maximum likelihood (ML) estimation is developed primarily for multivariable dynamic systems. The algorithm relies on a new optimization method referred to as a modified Newton-Raphson with estimated sensitivities (MNRES). The method determines sensitivities by using slope information from local surface approximations of each output variable in parameter space. The fitted surface allows sensitivity information to be updated at each iteration with a significant reduction in computational effort compared with integrating the analytically determined sensitivity equations or using a finite-difference method. Different surface-fitting methods are discussed and demonstrated. Aircraft estimation problems are solved by using both simulated and real-flight data to compare MNRES with commonly used methods; in these solutions MNRES is found to be equally accurate and substantially faster. MNRES eliminates the need to derive sensitivity equations, thus producing a more generally applicable algorithm.

Murphy, P. C.↗

NASA NTRS2013

Mars Entry Atmospheric Data System Trajectory Reconstruction Algorithms and Flight Results

The Mars Entry Atmospheric Data System is a part of the Mars Science Laboratory, Entry, Descent, and Landing Instrumentation project. These sensors are a system of seven pressure transducers linked to ports on the entry vehicle forebody to record the pressure distribution during atmospheric entry. These measured surface pressures are used to generate estimates of atmospheric quantities based on modeled surface pressure distributions. Specifically, angle of attack, angle of sideslip, dynamic pressure, Mach number, and freestream atmospheric properties are reconstructed from the measured pressures. Such data allows for the aerodynamics to become decoupled from the assumed atmospheric properties, allowing for enhanced trajectory reconstruction and performance analysis as well as an aerodynamic reconstruction, which has not been possible in past Mars entry reconstructions. This paper provides details of the data processing algorithms that are utilized for this purpose. The data processing algorithms include two approaches that have commonly been utilized in past planetary entry trajectory reconstruction, and a new approach for this application that makes use of the pressure measurements. The paper describes assessments of data quality and preprocessing, and results of the flight data reduction from atmospheric entry, which occurred on August 5th, 2012.

Karlgaard, Christopher D.↗

NASA NTRS1986

A data reduction package for multiple object spectroscopy

Experience with fiber-optic spectrometers has demonstrated improvements in observing efficiency for clusters of 30 or more objects that must in turn be matched by data reduction capability increases. The Medusa Automatic Reduction System reduces data generated by multiobject spectrometers in the form of two-dimensional images containing 44 to 66 individual spectra, using both software and hardware improvements to efficiently extract the one-dimensional spectra. Attention is given to the ridge-finding algorithm for automatic location of the spectra in the CCD frame. A simultaneous extraction of calibration frames allows an automatic wavelength calibration routine to determine dispersion curves, and both line measurements and cross-correlation techniques are used to determine galaxy redshifts.

Hill, J. M.↗

DOE OSTI2026

Solar-Tracker Diffuse-Response Algorithm for Balancing Energy Gain and Mechanical Wear in Photovoltaic Systems

Single-axis solar tracking maximizes photovoltaic energy production under clear-sky conditions; however, its effectiveness decreases under cloudy and overcast skies, where diffuse irradiance dominates and the optimal module orientation changes. Conventional tracking algorithms either neglect sky conditions or rely on simplified diffuse-response strategies that may trigger frequent tracker repositioning under variable cloud cover, leading to increased mechanical wear with marginal energy gains. This work proposes an enhanced diffuse-response tracking algorithm that explicitly accounts for both the intensity and temporal persistence of cloudiness. By requiring overcast conditions to persist for a minimum duration before reorienting the tracker to a diffuse-stow position, the proposed approach reduces unnecessary movements while preserving the benefits of diffuse-response operation. The algorithm is evaluated through numerical simulations based on historical meteorological data and validated using field measurements on monofacial and bifacial photovoltaic strings. The results show that the proposed strategy reduces excess tracker movement from 114% to 0.16% while maintaining nearly the same energy yield. Compared to a conventional diffuse-response algorithm, the associated energy reduction is minimal (≈0.17%) relative to the ≈0.37% yield gain observed at the studied location. These findings demonstrate that incorporating cloudiness duration enables a practical compromise between energy performance and tracker durability, particularly for monofacial photovoltaic systems.

14 SOLAR ENERGY↗

NASA NTRS1999

Flight Test of an Adaptive Configuration Optimization System for Transport Aircraft

A NASA Dryden Flight Research Center program explores the practical application of real-time adaptive configuration optimization for enhanced transport performance on an L-1011 aircraft. This approach is based on calculation of incremental drag from forced-response, symmetric, outboard aileron maneuvers. In real-time operation, the symmetric outboard aileron deflection is directly optimized, and the horizontal stabilator and angle of attack are indirectly optimized. A flight experiment has been conducted from an onboard research engineering test station, and flight research results are presented herein. The optimization system has demonstrated the capability of determining the minimum drag configuration of the aircraft in real time. The drag-minimization algorithm is capable of identifying drag to approximately a one-drag-count level. Optimizing the symmetric outboard aileron position realizes a drag reduction of 2-3 drag counts (approximately 1 percent). Algorithm analysis of maneuvers indicate that two-sided raised-cosine maneuvers improve definition of the symmetric outboard aileron drag effect, thereby improving analysis results and consistency. Ramp maneuvers provide a more even distribution of data collection as a function of excitation deflection than raised-cosine maneuvers provide. A commercial operational system would require airdata calculations and normal output of current inertial navigation systems; engine pressure ratio measurements would be optional.

Gilyard, Glenn B.↗

DOE OSTI2026

Long-term stabilization of intensity-difference squeezing from four-wave mixing in rubidium vapor

Generation of quantum states of light through off-resonance four-wave mixing in rubidium vapor is a straightforward and well-established technique. However, achieving a sufficiently high and consistent level of intensity difference squeezing (IDS) and intrinsic quantum correlations between photon pairs over an extended timescale, necessary for quantum-light-based nonlinear optical spectroscopy and microscopy, remains challenging and largely unexplored. Here, we report a simple stabilization method combining an active periodic laser frequency retuning with an automatic control algorithm based on quantitatively assessing the factors affecting the IDS level and squeezed light intensity. Validation of our method was performed by acquiring data over a 5-hour period, and the results demonstrate a remarkably stable squeezing level of -7.8 dB in combination with a > 4× reduction of the standard deviation of IDS level from 0.46 dB to 0.10 dB. The achieved stabilization further enables us to quantitatively assess the IDS reduction due to the scattering in a polystyrene bead suspension as a function of sample transmission. Our approach should enable a variety of applications requiring an extended squeezing stability over multiple hours, especially for those following biological processes and chemical reactions in real time.

Allen, Christian Harry [Oak Ridge National Laborat↗

NASA NTRS1999

Data Assimilation in the Presence of Forecast Bias: The GEOS Moisture Analysis

We describe the application of the unbiased sequential analysis algorithm developed by Dee and da Silva (1998) to the GEOS DAS moisture analysis. The algorithm estimates the persistent component of model error using rawinsonde observations and adjusts the first-guess moisture field accordingly. Results of two seasonal data assimilation cycles show that moisture analysis bias is almost completely eliminated in all observed regions. The improved analyses cause a sizable reduction in the 6h-forecast bias and a marginal improvement in the error standard deviations.

Dee, Dick P.↗

NASA NTRS1983

Algorithm development for the control design of flexible structures

The critical problems associated with the control of highly damped flexible structures are outlined. The practical problems include: high performance; assembly in space, configuration changes; on-line controller software design; and lack of test data. Underlying all of these problems is the central problem of modeling errors. To justify the expense of a space structure, the performance requirements will necessarily be very severe. On the other hand, the absence of economical tests precludes the availability of reliable data before flight. A design algorithm is offered which: (1) provides damping for a larger number of modes than the optimal attitude controller controls; (2) coordinates the rate of feedback design with the attitude control design by use of a similar cost function; and (3) provides model reduction and controller reduction decisions which are systematically connected to the mathematical statement of the control objectives and the disturbance models.

Skelton, R. E.↗

NASA NTRSDate not supplied

Jet Noise Prediction for Chevron Nozzle Flows with Wall-Modeled Large-Eddy Simulation

This paper presents results from ongoing research on jet noise prediction with wall-modeled large eddy simulations (WMLES) performed with the LAVA computational framework. In particular the focus of this study is on mixing enhancements from a single stream chevron nozzle at Reynolds number of 1×10(exp 6). Although the concept of chevron nozzles to reduce jet noise is not new, our understanding of its impact on the overall noise is still not well understood. As a first step towards predicting noise reduction due to mixing enhancement concepts from first principles with WMLES, we simulate the noise generated by a single stream chevron nozzle SMC001 as well as its equivalent axisymmetric round jet nozzle SMC000. Detailed comparisons are made with a dedicated experiment conducted at NASA’s Glenn Research Center and good agreement was achieved. Two different approaches to introduce a turbulent boundary layer were compared but show no major impact on the results. This is especially important given future work were multi-stream nozzles are considered and extended costs of resolving the internal BL would have a bigger cost impact. A permeable Ffowcs Williams Hawkings (FWH) surface enclosing the jet is used to predict far-field noise from the simulated flow-field and excellent comparison to microphone array measurements is achieved within the resolved frequency band. Sensitivity of far-field noise predictions to grid resolution is systematically documented. Near-field comparisons to PIV data shows great agreement for both velocity and normal stresses, however a systematic TKE overshoot at the nozzle exit is seen in the shear-layer. The paper shows a cost reduction of an order of magnitude compared to an earlier study of this configuration due to algorithmic and software improvements and demonstrates that WMLES can be used as a cost-competitive approach for jet noise predictions.

CST↗

NASA NTRSDate not supplied

Proposition to Optimize Fire Stations for Wildfires

The rising trend in wildfire occurrence and severity has put a strain on wildfire management organizations by spreading out limited resources to meet increasing demand. There has been extensive prior research and data collection to determine the areas of highest risk and to predict regional wildfire damages based on historical trends. Our group aims to utilize this data to best determine fire station placement, optimizing where resources are allocated to reduce the time and investment needed to effectively mitigate wildfires. Using existing research, we are able to calculate optimal fire station locations by utilizing a single-objective facility location problem algorithm incentivized for cost reduction. Further work would be needed to refine the algorithm to accommodate for more realistic factors including access to water and roadways as well as better accounting for the costs involved, but our work serves as a proof of concept and lays the foundation for future research. Implementation of this algorithm would allow public fire planning agencies (such as CAL FIRE and Forest Service) to shift resources to where they would be most effective.

Wildfires↗

DOE OSTI2024

Reduce-Order Modeling of Multigroup Neutron Cross Sections for High-Temperature Gas-cooled Reactors

Deterministic neutronics calculations rely on multigroup neutron cross section libraries, which usually consists of a database of tabulated values, used to calculate the cross sections through multivariate linear interpolation. However, interpolation of the multidimensional cross section data becomes memory inefficient and time consuming as the number of tabulations increases, significantly slowing down the neutronics calculation, especially in the case of micro cross section libraries where every isotope (on the order of hundreds) has its own set of specific reactions and cross sections. To address this challenge, this work constructs efficient and robust reduced-order models (ROMs) of the multi-group cross sections to support the Griffin simulation of high-temperature gas-cooled reactors (HTGRs). The first part of the study investigates the linearity of the multi-group cross section data across isotopes, reaction types and energy groups on pre-generated datasets for the purpose of dimensionality reduction. Secondly, a down-selection of ROM techniques is presented on representative classical machine learning (ML) techniques, including variants of linear regression, kernel-based methods, tree-based algorithms, and artificial neural networks. The selection criteria jointly consider the memory efficiency, predictive accuracy, prediction speed, and scalability in comparison to the multidimensional interpolation. Among all the ML techniques, deep neural networks (DNNs) have proven to be the best selection with sufficient accuracy, high robustness, good memory efficiency, great scalability, and superior flexibility. DNNs for have been trained for all isotopes in this work and systematic Griffin testing is ongoing at this moment to ensure the feasibility of this ROM technique for cross section predictions.

42 - ENGINEERING↗

DOE OSTI2024

Reduced-Order Modeling of Multigroup Neutron Cross Sections for High-Temperature Gas-cooled Reactors

Abstract – Deterministic neutronics calculations rely on multigroup neutron cross section libraries, which consist of databases of tabulated values, used to calculate the neutron cross sections through multivariate linear interpolation. However, interpolation of the multidimensional cross section data becomes memory inefficient and time consuming as the number of tabulations increases, significantly slowing down the neutronics calculation, especially in the case of microscopic cross section libraries where every isotope (on the order of hundreds) has its own set of specific reactions and cross sections. In order to address this challenge, this work constructs efficient and robust reduced-order models (ROMs) of the multi-group cross sections to support the Griffin simulation of high-temperature gas-cooled reactors (HTGRs). The first part of the study investigates the linearity of the multigroup cross section data across isotopes, reaction types, and energy groups on pre-generated datasets for the purpose of dimensionality reduction. Secondly, a down-selection of ROM techniques is presented on representative classical machine learning (ML) techniques, including variants of linear regression, kernel-based methods, tree-based algorithms, and artificial neural networks. The selection criteria jointly consider the memory efficiency, predictive accuracy, prediction speed, and scalability in comparison to the multidimensional interpolation. Among all the ML techniques, deep neural networks (DNNs) have proven to be the best selection with sufficient accuracy, high robustness, good memory efficiency, great scalability, and superior flexibility. DNNs have been trained for all isotopes in this work and systematic Griffin testing is ongoing to ensure the feasibility of this ROM technique for predicting cross section and reducing memory requirements without a significant sacrifice in computational performance.

42 - ENGINEERING↗

DOE OSTI2024

Advanced Cross Section Library Generation using Reduced Order Models

Deterministic neutronics calculations rely on multigroup neutron cross section libraries, which consist of databases of tabulated values, used to calculate the neutron cross sections through multivariate linear interpolation. However, interpolation of the multidimensional cross section data becomes memory inefficient and time consuming as the number of tabulations increases, significantly slowing down the neutronics calculation, especially in the case of microscopic cross section libraries where every isotope (on the order of hundreds) has its own set of specific reactions and cross sections. In order to address this challenge, this work constructs efficient and robust reduced-order models (ROMs) of the multi-group cross sections to support the Griffin simulation of high-temperature gas-cooled reactors (HTGRs). The first part of the study investigates the linearity of the multigroup cross section data across isotopes, reaction types, and energy groups on pre-generated datasets for the purpose of dimensionality reduction. Secondly, a down-selection of ROM techniques is presented on representative classical machine learning (ML) techniques, including variants of linear regression, kernel-based methods, tree-based algorithms, and artificial neural networks. The selection criteria jointly consider the memory efficiency, predictive accuracy, prediction speed, and scalability in comparison to the multidimensional interpolation. Among all the ML techniques, deep neural networks (DNNs) have proven to be the best selection with sufficient accuracy, high robustness, good memory efficiency, great scalability, and superior flexibility. DNNs have been trained for all isotopes in this work and systematic Griffin testing is ongoing to ensure the feasibility of this ROM technique for predicting cross section and reducing memory requirements without a significant sacrifice in computational performance.

42 - ENGINEERING↗

DOE OSTI2024

Extended dynamic mode decomposition for model reduction in fluid dynamics simulations

High computational cost and storage/memory requirements of fluid dynamics simulations constrain their usefulness as a predictive tool. Reduced-order models (ROMs) provide a viable solution to this challenge by extracting the key underlying dynamics of a complex system directly from data. We investigate the efficacy and robustness of an extended dynamic mode decomposition (xDMD) algorithm in constructing ROMs of three-dimensional cardiovascular computations. Focusing on the ROMs' accuracy in representation and interpolation, we relate these metrics to the truncation rank of singular value decomposition, which underpins xDMD and other approaches to ROM construction. Our key innovation is to relate the truncation rank to the singular values of the original flow problem. This result establishes a priori guidelines for the xDMD deployment and its likely success as a means of data compression and reconstruction of the system's dynamics from dominant spatiotemporal structures present in the data.

Mechanics↗

DOE OSTI2025

Machine learning-assisted identification of potential sources of bias in measurements of prompt-fission neutron spectra

Unrecognized sources of uncertainty (USU) can bias the reported mean and/or covariance of experimental nuclear data. These biases, in turn, can propagate through evaluated nuclear data to application simulations or may poorly inform nuclear theory that is fitted to the experimental data. Such unknown sources of bias must be tied to the inherent physical constituents of the measurements such as the characteristics of a detector response or a background reduction technique. Here, in this article, a sparse Bayesian learning model is used to support experts in their efforts to identify and characterize USU in experimental prompt fission neutron spectra (PFNS) for spontaneous fissioning of 252 Cf by linking observed biases to features of the measurement system. Three different bias components were found. The first acts as a verification case for the algorithm as it identifies a bias coming from a well-known source related to the use of 6 Li in the neutron detection system. The second two cases demonstrate how this method can benefit the evaluation of experimental nuclear data by identifying, quantifying, and relating unknown biases to potential causes.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2025

Accelerating multilevel Markov Chain Monte Carlo using machine learning models

Here, this work presents an efficient approach for accelerating multilevel Markov Chain Monte Carlo (MCMC) sampling for large-scale problems using low-fidelity machine learning models. While conventional techniques for large-scale Bayesian inference often substitute computationally expensive high-fidelity models with machine learning models, thereby introducing approximation errors, our approach offers a computationally efficient alternative by augmenting high-fidelity models with low-fidelity ones within a hierarchical framework. The multilevel approach utilizes the low-fidelity machine learning model (MLM) for inexpensive evaluation of proposed samples thereby improving the acceptance of samples by the high-fidelity model. The hierarchy in our multilevel algorithm is derived from geometric multigrid hierarchy. We utilize an MLM to accelerate the coarse level sampling. Training machine learning model for the coarsest level significantly reduces the computational cost associated with generating training data and training the model. We present an MCMC algorithm to accelerate the coarsest level sampling using MLM and account for the approximation error introduced. We provide theoretical proofs of detailed balance and demonstrate that our multilevel approach constitutes a consistent MCMC algorithm. Additionally, we derive the expression for cost reduction due to machine learning model to facilitate cost analysis of the hierarchical sampling algorithm. Our technique is demonstrated on a standard benchmark inference problem in groundwater flow, where we estimate the probability density of a quantity of interest using a four-level MCMC algorithm. Our proposed algorithm accelerates multilevel sampling by a factor of two while achieving similar accuracy compared to sampling using the standard multilevel algorithm.

97 MATHEMATICS AND COMPUTING↗