Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Adaptive compression”

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

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

At least 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Dynamical Low-Rank Compression of Neural Networks with Robustness under Adversarial Attacks

Deployment of neural networks on resource-constrained devices demands models that are both compact and robust to adversarial inputs. However, compression and adversarial robustness often conflict. In this work, we introduce a dynamical low-rank training scheme enhanced with a novel spectral regularizer that controls the condition number of the low-rank core in each layer. This approach mitigates the sensitivity of compressed models to adversarial perturbations without sacrificing clean accuracy. The method is model- and data-agnostic, computationally efficient, and supports rank adaptivity to automatically compress the network at hand. Extensive experiments across standard architectures, datasets, and adversarial attacks show the regularized networks can achieve over 94 compression while recovering or improving adversarial accuracy relative to uncompressed baselines.

Schotthoefer, Steffen [ORNL] (ORCID:00000002156965↗
DOE OSTI2020

Smart Scattering Scanning Near-Field Optical Microscopy

Scattering scanning near-field optical microscopy (s-SNOM) provides spectroscopic imaging from molecular to quantum materials with few nanometer deep subdiffraction limited spatial resolution. However, in its conventional implementation s-SNOM is slow to effectively acquire a series of spatio-spectral images, especially with large fields of view. This problem is further exacerbated for weak resonance contrast or when using light sources with limited spectral irradiance. Indeed, the generally limited signal-to-noise ratio prevents sampling a weak signal at the Nyquist sampling rate. Here, we demonstrate how acquisition time and sampling rate can be significantly reduced by using compressed sampling, matrix completion, and adaptive random sampling, while maintaining or even enhancing the physical or chemical image content. We use fully sampled real data sets of molecular, biological, and quantum materials as ground-truth physical data and show how deep under-sampling with a corresponding reduction of acquisition time by 1 order of magnitude or more retains the core s-SNOM image information. We demonstrate that a sampling rate of up to 6× smaller than the Nyquist criterion can be applied, which would provide a 30-fold reduction in the data required under typical experimental conditions. Furthermore, our smart s-SNOM approach is generally applicable and provides systematic full spatio-spectral s-SNOM imaging with a large field of view at high spectral resolution and reduced acquisition time.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Scientific Data Compression for Large Scale Computational Fluid Dynamics (CFD) Simulations

This Cooperative Research and Development Agreement (CRADA) between Oak Ridge National Laboratory (ORNL) and General Electric (GE) investigated methods for reducing the size of large computational fluid dynamics (CFD) simulation datasets using scientific data compression techniques. The work focused on adapting the MultiGrid Adaptive Reduction of Data (MGARD) compression framework and integrating it with high-performance I/O and visualization tools used in CFD workflows. MGARD uses hierarchical multilevel decomposition to enable error-controlled compression of floating-point scientific data while preserving quantities of interest. During the project, MGARD compression was integrated with the ADIOS I/O framework and visualization tools such as ParaView to enable efficient storage, transfer, and analysis of simulation data. The collaboration also explored approaches for improving compression performance for CFD data defined on unstructured meshes. Results demonstrate that scientific data compression can significantly reduce storage requirements and improve data management for large-scale CFD simulations.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2007

A Low-Complexity Circuit for On-Sensor Concurrent A/D Conversion and Compression

A low-complexity circuit for on-sensor compression is presented. The proposed circuit achieves complexity savings by combining a single-slope analog-to-digital converter with a Golomb-Rice entropy encoder and by implementing a low-complexity adaptation rule. The adaptation rule monitors the output codewords and minimizes their length by incrementing or decrementing the value of the Golomb-Rice coding parameter k. Its hardware implementation is one order of magnitude lower than existing adaptive algorithms. The compression circuit has been fabricated using a 0.35 micrometers CMOS technology and occupies an area of 0.0918 mm2. Test measurements confirm the validity of the design

Leon-Salas, Walter D.↗
NASA NTRS1987

Finite element methods for compressible flows

The problems of mesh generation and developing effective algorithms for the solution of the equations of compressible flow on unstructured meshes are discussed. Adaptive mesh refinement methods can be implemented in a straightforward manner. Possible adaptive strategies are examined. A finite element method adapted to problems involving high speed compressible flow is described. The adaptive mesh regeneration procedure appears to offer the possibility of large computational savings in three dimensional flow computation.

Morgan, K.↗
NASA NTRS1995

Image data compression having minimum perceptual error

A method for performing image compression that eliminates redundant and invisible image components is described. The image compression uses a Discrete Cosine Transform (DCT) and each DCT coefficient yielded by the transform is quantized by an entry in a quantization matrix which determines the perceived image quality and the bit rate of the image being compressed. The present invention adapts or customizes the quantization matrix to the image being compressed. The quantization matrix comprises visual masking by luminance and contrast techniques and by an error pooling technique all resulting in a minimum perceptual error for any given bit rate, or minimum bit rate for a given perceptual error.

Watson, Andrew B.↗
NASA NTRS2022

The Effects of Co-cured Versus Precured Carbon/Epoxy Face Sheets on the Modulus and Compression After Impact Strength of Honeycomb Sandwich Structure

This study presents experimental results of compression after impact (CAI) and stiffness testing of aluminum honeycomb core sandwich structure with quasi-isotropic face sheets that have either been co-cured or precured (and then bonded to the core in a secondary operation). The co-cured face sheets contain some ply waviness whereas the precured face sheets did not. The effect of this waviness was ascertained for two important design parameters for the Space Launch System (SLS) Payload Adapter Fitting (PAF); compressive modulus and CAI strength. It was found that there was no effect on the modulus between co-cured and precured face sheets using the material and processes in this study. In addition, there was no difference in CAI strength behavior between the co-cured and precured face sheet.

sandwich structure↗
DOE OSTI2021

Optimizing Error-Bounded Lossy Compression for Scientific Data on GPUs

Error-bounded lossy compression is a critical technique for significantly reducing scientific data volumes. With ever-emerging heterogeneous high-performance computing (HPC) architecture, GPU-accelerated error-bounded compressors (such as CUSZ and cuZFP) have been developed. However, they suffer from either low performance or low compression ratios. To this end, we propose CUSZ+ to target both high compression ratios and throughputs. We identify that data sparsity and data smoothness are key factors for high compression throughputs. Our key contributions in this work are fourfold: (1) We propose an efficient compression workflow to adaptively perform run-length encoding and/or variable-length encoding. (2) We derive Lorenzo reconstruction in decompression as multidimensional partial-sum computation and propose a fine-grained Lorenzo reconstruction algorithm for GPU architectures. (3) We carefully optimize each of CUSZ kernels by leveraging state-of-the-art CUDA parallel primitives. (4) We evaluate CUSZ+ using seven real-world HPC application datasets on V100 and A100 GPUs. Experiments show CUSZ+ improves the compression throughputs and ratios by up to 18.4x and 5.3x, respectively, over CUSZ on the tested datasets.

Tian, Jiannan↗
DOE OSTI2021

Bayesian tensorized neural networks with automatic rank selection

Tensor decomposition is an effective approach to compress over-parameterized neural networks and to enable their deployment on resource-constrained hardware platforms. However, directly applying tensor compression in the training process is a challenging task due to the difficulty of choosing a proper tensor rank. In order to address this challenge, this paper proposes a low-rank Bayesian tensorized neural network. Our Bayesian method performs automatic model compression via an adaptive tensor rank determination. We also present approaches for posterior density calculation and maximum a posteriori (MAP) estimation for the end-to-end training of our tensorized neural network. Here, we provide experimental validation on a two-layer fully connected neural network, a 6-layer CNN and a 110-layer residual neural network where our work produces 7.4x to 137x more compact neural networks directly from the training while achieving high prediction accuracy.

Low-rank tensor↗
DOE OSTI2025

Hybrid learning techniques for scientific data reduction with performance guarantees

The research initiatives supported by the U.S. Department of Energy (DOE) Grant DE-SC0022265 are fundamentally aimed at pioneering advanced machine learning (ML) techniques for scientific data compression within high-performance computing (HPC) environments. This comprehensive body of work addresses the critical challenge posed by the exponential growth of data generated by scientific simulations in domains such as fusion energy, climate modeling, and computational fluid dynamics (CFD). A core objective is to develop compression algorithms that achieve substantial data reduction—often by orders of magnitude—while rigorously ensuring the fidelity of both the primary data (PD) and scientifically crucial derived quantities of interest (QoI). The methodologies deployed under this grant integrate sophisticated deep learning architectures, prominently featuring autoencoders, advanced generative models like conditional diffusion, and hybrid learning techniques. Key innovations include the development of Guaranteed Autoencoders (GAE) and the Guaranteed Conditional Diffusion with Tensor Correction (GCDTC) framework, which provide explicit, instance-level error bounds on reconstructed data. Furthermore, specialized strategies such as nonlinear constraint satisfaction are employed to preserve the integrity of QoI, a vital requirement for the trustworthiness of downstream scientific analyses. This research also focuses on the design and implementation of scalable, GPU-accelerated software pipelines that seamlessly integrate into existing HPC workflows, ensuring both computational efficiency and practical applicability. The CAESAR framework, for example, unifies foundation and generative models to create an adaptive and efficient compression solution for spatio-temporal scientific data. Collectively, these efforts represent a significant advancement in mitigating the scientific data deluge, enabling more effective data management, accelerated scientific discovery, and optimized utilization of HPC resources.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Final report- UFL - RAPIDS2: A SciDAC Institute for Computer Science, Data, and Artificial Intelligence

The research initiatives supported by the U.S. Department of Energy (DOE) Grant DE-SC0022265 are fundamentally aimed at pioneering advanced machine learning (ML) techniques for scientific data compression within high-performance computing (HPC) environments. This comprehensive body of work addresses the critical challenge posed by the exponential growth of data generated by scientific simulations in domains such as fusion energy, climate modeling, and computational fluid dynamics (CFD). A core objective is to develop compression algorithms that achieve substantial data reduction—often by orders of magnitude—while rigorously ensuring the fidelity of both the primary data (PD) and scientifically crucial derived quantities of interest (QoI). The methodologies deployed under this grant integrate sophisticated deep learning architectures, prominently featuring autoencoders, advanced generative models like conditional diffusion, and hybrid learning techniques. Key innovations include the development of Guaranteed Autoencoders (GAE) and the Guaranteed Conditional Diffusion with Tensor Correction (GCDTC) framework, which provide explicit, instance-level error bounds on reconstructed data. Furthermore, specialized strategies such as nonlinear constraint satisfaction are employed to preserve the integrity of QoI, a vital requirement for the trustworthiness of downstream scientific analyses. This research also focuses on the design and implementation of scalable, GPU-accelerated software pipelines that seamlessly integrate into existing HPC workflows, ensuring both computational efficiency and practical applicability. The CAESAR framework, for example, unifies foundation and generative models to create an adaptive and efficient compression solution for spatio-temporal scientific data. Collectively, these efforts represent a significant advancement in mitigating the scientific data deluge, enabling more effective data management, accelerated scientific discovery, and optimized utilization of HPC resources.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2004

Real time on-chip sequential adaptive principal component analysis for data feature extraction and image compression

In this paper, we present a new, simple, and optimized hardware architecture sequential learning technique for adaptive Principle Component Analysis (PCA) which will help optimize the hardware implementation in VLSI and to overcome the difficulties of the traditional gradient descent in learning convergence and hardware implementation.

PCA feature extraction image compression real time↗
NASA NTRS1989

Finite element analysis of high speed compressible flows using mesh refinement/movement procedures

An adaptive mesh refinement procedure for analyzing high-speed inviscid and viscous compressible flows is described. The adaptation procedure which uses both quadrilateral and triangular elements is implemented with an explicit finite element formulation. Elements in regions of strong and weak gradients are refined or coarsened based on inviscid and viscous indicators. Nodal locations are also adaptively moved to better resolve flow features. The effectiveness of the finite element procedure is demonstrated by modeling flows with complex shock structure and viscous-inviscid interactions. Numerical results are compared with experimental data.

Ramakrishnan, R.↗
NASA NTRS1976

A real-time adaptive Hadamard transform video compressor

An adaptive video compressor was designed as part of the CTS Digital Video Curriculum Sharing Experiment. The compressor was constructed using field-to-field differencing on fields processed using a Hadamard transform method. The spatial resolution was improved using a fixed-rate, three-mode adaptive system to compress each field.

Jones, H. W., Jr.↗
NASA NTRS1991

Generalized multi-point inverse airfoil design

In a rather general sense, inverse airfoil design can be taken to mean the problem of specifying a desired set of airfoil characteristics, such as the airfoil maximum thickness ratio, pitching moment, part of the velocity distribution or boundary-layer development, etc., then from this information determine the corresponding airfoil shape. This paper presents a method which approaches the design problem from this perspective. In particular, the airfoil is divided into segments along which, together with the design conditions, either the velocity distribution or boundary-layer development may be prescribed. In addition to these local desired distributions, single parameters like the airfoil thickness can be specified. The problem of finding the airfoil shape is determined by coupling an incompressible, inviscid, inverse airfoil design method with a direct integral boundary-layer analysis method and solving the resulting nonlinear equations via a multidimensional Newton iteration technique. The approach is fast and easily allows for interactive design. It is also flexible and could be adapted to solving compressible, inverse airfoil design problems.

Selig, Michael S.↗
NASA NTRS2021

Scheduling and Operations of the Orbiting Carbon Observatory-3 Mission

This paper describes the development and use of an automated scheduling system for the National Aeronautics and Space Administration’s (NASA) Orbiting Carbon Observatory-3 (OCO-3) Mission. OCO-3 measures atmospheric carbon dioxide from space. Made from the spare instrument built as a backup to the Orbiting Carbon Observatory-2 (OCO-2), OCO-3 extends the rich set of data collected by OCO-2. OCO-3 is outfitted with an agile Pointing Mirror Assembly (PMA) that allows for more detailed types of observations and rapid mode transitions. The mission uses an adaptation of the Compressed Large-scale Activity Scheduling and Planning (CLASP) system for scheduling nominal operations, as well as a separate automated scheduling system developed for scheduling observations for the calibration of the PMA. CLASP is used to schedule the four types of observational modes: Nadir, Glint, Target, and Snapshot Area Map. OCO-3 has a variety of complex mission-specific geometric constraints that were incorporated into CLASP to produce schedules that ensure instrument safety.

Moy, Alan↗
NASA NTRS2021

Scheduling and Operations of the ECOSTRESS Mission

This paper describes the development and use of an automated scheduling system for the National Aeronautics and Space Administration’s (NASA) ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) mission. Key to the success of the ECOSTRESS mission has been the use of automated scheduling in mission analysis pre-launch, and in successful operations where automated scheduling was deployed to address several operational challenges. ECOSTRESS uses an adaptation of the Compressed Large-scale Activity Scheduling and Planning (CLASP) system to automatically select science observations respecting area and point target priorities as well as visibility, illumination, onboard storage, and radiation constraints to satisfy high-level prioritized science campaigns. The ECOSTRESS scheduler was used pre-launch to predict the effectiveness of alternative formulations of science campaign definitions accounting for the impact of data volume, keepout, and orbit/illumination/visibility constraints to derive the initial operational science campaign definitions and priorities. The scheduler was then used after instrument checkout for operations. ECOSTRESS has faced multiple operational challenges relating to instrument firmware and hardware, and the scheduler has been updated several times to address these challenges. The instrument Mass Storage Units (MSUs) had operational issues, requiring the scheduler to plan for and schedule commands to handle intricacies of data management. After many months of operations, both MSUs on the instrument became non-functioning and the firmware of the instrument was updated to bypass the MSUs. A further update to the ECOSTRESS scheduler enabled the scheduler to operate in this new operations mode. The ECOSTRESS scheduler has also been updated to improve handling of along-track uncertainty inherent in International Space Station operations. The flexibility and ease of updating of the automated scheduler has been a significant contributor to successful operations of the ECOSTRESS mission.

Padams, Jordan↗