Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Edge computing”

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 757 records · Page 42

Single-Net Continual Learning with Progressive Segmented Training

There is an increasing need of continual learning in dynamic systems, such as the self-driving vehicle, the surveillance drone, and the robotic system. Such a system requires learning from the data stream, training the model to preserve previous information and adapt to a new task, and generating a single-headed vector for future inference. Different from previous approaches with dynamic structures, this work focuses on a single network and model segmentation to prevent catastrophic forgetting. Leveraging the redundant capacity of a single network, model parameters for each task are separated into two groups: one important group which is frozen to preserve current knowledge, and secondary group to be saved (not pruned) for a future learning. A fixed-size memory containing a small amount of previously seen data is further adopted to assist the training. Without additional regularization, the simple yet effective approach of Progressive Segmented Training (PST) successfully incorporates multiple tasks and achieves the state-of-the-art accuracy in the single-head evaluation on CIFAR-10 and CIFAR-100 datasets. Moreover, the segmented training significantly improves computation efficiency in continual learning at the edge.

Du, Xiaocong↗

A Blockchain-Enabled Sustainable Safety Management Framework for Connected Vehicles

The notion of an intelligent transportation system (ITS) aims to boost the performance of transportation networks, which has gained more and more traction in both academic and commercial circles. ITS is a constantly evolving vision that combines cutting-edge transportation approaches with new information, communication, computers, and other technology. ITS should discover consequence routes to enhance the sustainability, safety, and trustworthiness of the entire transportation system utilizing emerging technologies. In this paper, a sustainable safety management framework for connected vehicles is proposed by integrating blockchain. It introduces smart transportation equipment called an AI-enabled vehicle smart device (AVSD) for vehicular communications. AVSD can reduce energy consumption by decreasing the computational costs in vehicular communications. Smart contracts are used to identify vehicles automatically and establish secure communication among vehicles and emergency service stations (ESSs) like hospitals, police stations, and fire stations. The experiment results show that the proposed framework provides a communication environment for sustainable safety and security using the introduced smart transportation device. In conclusion, the proposed blockchain-enabled sustainable safety management framework has the potential to improve safety and sustainability in the transportation industry by creating a secure, decentralized, and transparent platform for managing safety data and promoting safe and sustainable driving behaviors.

42 ENGINEERING↗

Interdisciplinary Dialogue for Education, Collaboration, and Innovation: Intelligent Biology and Medicine In and Beyond 2013

The 2013 International Conference on Intelligent Biology and Medicine (ICIBM 2013) was held on August 11-13, 2013 in Nashville, Tennessee, USA. The conference included six scientific sessions, two tutorial sessions, one workshop, two poster sessions, and four keynote presentations that covered cutting-edge research topics in bioinformatics, systems biology, computational medicine, and intelligent computing. Here, we present a summary of the conference and an editorial report of the supplements to BMC Genomics and BMC Systems Biology that include 19 research papers selected from ICIBM 2013.

bioinformatics↗

Nanoparticle self-assemblies with modern complexity

Thanks to decades of tireless efforts, nanoparticle assemblies have reached at an extremely high level of controllability, sophistication, and complexity, with new insights provided by integration with graph theory, cutting-edge characterization, and machine learning (ML)-based computation and modeling, as well as with ever-diversifying applications in energy, catalysis, biomedicine, optics, electronics, magnetics, organic biosynthesis, and quantum technology. Nanoparticle assemblies can be crystalline, known as superlattices or supracrystals. Their assembly entails a transition from disorder—dispersed nanoparticles—to order, which can be achieved through classical nucleation pathways or nonclassical pathways via prenucleation precursors or particle aggregation. Further, the periodic lattices allow facile manipulations of electrons, phonons, photons, and even spins, leading to advanced device components and metamaterials. Meanwhile, aperiodic assemblies out of nanoparticles, such as gels, networks, and amorphous solids, also start to attract attentions. Despite the loss of periodicity, symmetry-lowering or symmetry-breaking three-dimensional (3D) structures emerge with unique properties, such as chiroptical activity, topological mechanical strength, and quantum entanglement. Real-space imaging such as electron microscopy and X-ray based tomography methods are utilized to characterize these complex structures, while mathematical tools such as graph theories are in need to describe such complex structures. This issue aims to provide a timely review of the efforts in this greatly broadened materials design space including experiment, simulation, theory, and applications. Nine top experts (and their teams) from four countries deliver six review papers, summarizing fundamental mechanistic understandings of nanoparticle assemblies, highlighted with the developments of state-of-the-art in situ characterization tools and ML-assisted reverse engineering, and newly emergent applications of nanoarchitectures.

36 MATERIALS SCIENCE↗

AToM (Advanced Tokamak Modeling)

This final report for the AToM project documents the contributions from General Atomics over the period 1 September 2014 to 31 August 2017. This includes significant progress on the development of tokamak integrated modeling frameworks (OMFIT and IPS), and development of new workflows for computing plasma profile (temperature, density, etc) evolution. Significant effort was also carried out in the high-performance computing arena, by adding the COGENT edge-simulation code to the AToM portfolio, and performing optimization of the CGYRO plasma turbulence code. Issues of data management and user outreach are also documented.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Large-Scale Trajectory Analysis via Feature Vectors

The explosion of both sensors and GPS-enabled devices has resulted in position/time data being the next big frontier for data analytics. However, many of the problems associated with large numbers of trajectories do not necessarily have an analog with many of the historic big-data applications such as text and image analysis. Modern trajectory analytics exploits much of the cutting-edge research in machine-learning, statistics, computational geometry and other disciplines. We will show that for doing trajectory analytics at scale, it is necessary to fundamentally change the way the information is represented through a feature-vector approach. We then demonstrate the ability to solve large trajectory analytics problems using this representation.

58 GEOSCIENCES↗

Bayesian chain graph models to characterize microbe-environment dynamics

Microbiome data require statistical models that can simultaneously decode microbes' reaction to the environment and interactions among microbes. While a multiresponse linear regression model seems like a straight-forward solution, we argue that treating it as a graphical model is problematic given that the regression coefficient matrix does not encode the conditional dependence structure between response and predictor nodes. This observation is especially important in biological settings when we have prior knowledge on the edges from specific experimental interventions that can only be properly encoded under a conditional dependence model. Here, we propose a chain graph model with two sets of nodes (predictors and responses) whose solution yields a graph with edges that indeed represent conditional dependence, thus agreeing with the experimenter's intuition on the average behavior of nodes under treatment. The solution to our model is sparse via the Bayesian linear regression (LASSO). In addition, we propose an adaptive extension so that different shrinkages can be applied to different edges to incorporate edge-specific prior knowledge. Our model is computationally inexpensive through an efficient Gibbs sampling algorithm and can account for binary, counting, and compositional responses via an appropriate hierarchical structure. We test the performance of our model in a variety of simulated datasets, thereby showing superior performance to state-of-the-art approaches. We further apply our model to human gut and soil microbial compositional datasets, and we highlight that CG-LASSO can estimate biologically meaningful network structures in the data.

compositional data↗

Wake curvature and trailing edge interaction effects in viscous flow over airfoils

A theory developed for analyzing viscous flows over airfoils at high Reynolds numbers is described. The theory includes a complete treatment of viscous interaction effects induced by the curved wake behind the airfoil and accounts for normal pressure gradients across the boundary layer in the trailing edge region. A brief description of a computer code that was developed to solve the extended viscous interaction equations is given. Comparisons of the theoretical results with wind tunnel data for two rear loaded airfoils at supercritical conditions are presented.

Melnik, R. E.↗

A comparison of cloud motion winds from ATS 6 images with coinciding SMS 1 winds

A methodology is developed for accurate measurement of cloud motion winds from the geosynchronous ATS 6 image data. Attitude changes between consecutive images (as a function of scan-line number) are accounted for in wind computations through measurement of the earth-edge displacements between the successive infrared images. Also, an image matching procedure is used to remove obvious and distracting image distortions. The availability of SMS imagery coinciding with ATS 6 imagery makes SMS an excellent reference against which the quality of ATS 6 winds can be tested. The resulting winds inferred from cloud displacement measurements taken from a sequence of the corrected images are found to agree better than 2 m/sec rms with winds measured from coincident SMS 1 imagery.

Kuhlow, W. W.↗

LANDSAT-D investigations in snow hydrology

Progress on the registration of TM data to digital topographic data; on comparison of TM, MSS and NOAA meteorological satellite data for snowcover mapping; and on radiative transfer models for atmospheric correction is reported. Some methods for analyzing spatial contiguity of snow within the snow covered area were selected. The methods are based on a two-channel version of the grey level co-occurence matrix, combined with edge detection derived from an algorithm for computing slopes and exposures from digital terrain data.

Dozier, J.↗

Evaluation of the solar disk sextant concept

In this paper the viability of the solar disk sextant concept is evaluated, the optimum parameters required to carry out solar variability studies, which are the mission objectives, are derived. The experimental environment is first discussed, followed by the application of the finite Fourier transform definition (FFTD) to the detector array data. The requirements on the optical system are studied next. A computer program was carried out simulating solar edge data and FFTD. From this study, it is concluded that the required accuracy of measurement may be reached using currently available detector array technology, a focal ratio of the optical system in excess of 90, and an entrance aperture of 22 cm. The guidance error must be small enough to require no more than a correction rate of 0.1 arcsec/sec. All these conditions are well within current technology.

Chiu, H.-Y.↗

Advanced propeller research

Resent results of aerodynamic and acoustic research on both single and counter-rotation propellers are reviewed. Data and analytical results are presented for three propellers: SR-7A, the single rotation design used in the NASA Propfan Test Assessment (PTA); and F7-A7, the 8+8 counterrotating design used in the proof-of-concept Unducted Fan (UDF) engine. In addition to propeller efficiencies, cruise and takeoff noise, and blade pressure data, off-design phenomena involving formation of leading edge vortices are described. Aerodynamic and acoustic computational results derived from three-dimensional Euler and acoustic radiation codes are presented. Research on unsteady flows, which are particularly important for understanding counterrotation interaction noise, unsteady loading effects on acoustics, and flutter or forced response is described. The first results of three-dimensional unsteady Euler solutions are illustrated for a single rotation propeller at an angle of attack and for a counterrotation propeller. Basic experimental and theoretical results from studies of the unsteady aerodynamics of oscillating cascades are outlined. Finally, advanced concepts involving swirl recovery vanes and ultra bypass ducted propellers are discussed.

Groeneweg, John F.↗

Tests on a sodium/Hastelloy X wing leading edge heat pipe for hypersonic vehicles

A full-scale, internally instrumented, sodium/Hastelloy X heat pipe was designed and fabricated for the wing leading edge of an advanced reentry vehicle. A computer model was developed to predict the heat pipe startup. The heat pipe was tested, and the results were used to validate the startup methodologies. In addition, valuable fabrication and test procedure lessons were learned.

Boman, B. L.↗

Numerical aspects of computing high Reynolds number flows on unstructured meshes

An edge-data structure describing a mesh edge-wise given the vertices of each edge and neighboring cell information is used developing algorithms for the Navier-Stokes equations on triangular meshes. Edge formulas for the Galerkin and finite-element discretization of gradient, divergence, Hessian, and Laplacian operators are derived. A simple edge formula is derived for the discretization of the Laplacian operator, where precise theoretical conditions for a discrete maximum principle can be ascertained. Practical issues associated with solving the Navier-Stokes equations on unstructured meshes are addressed, along with issues concerning the generation of highly stretched triangular meshes and the modeling of turbulence on unstructured meshes. A turbulence modeling strategy is proposed, and numerical results for a high-Reynolds-number flow about single- and multielement airfoils are discussed.

Barth, Timothy J.↗

Generalized conjugate-gradient methods for the Navier-Stokes equations

A generalized conjugate-gradient method is used to solve the two-dimensional, compressible Navier-Stokes equations of fluid flow. The equations are discretized with an implicit, upwind finite-volume formulation. Preconditioning techniques are incorporated into the new solver to accelerate convergence of the overall iterative method. The superiority of the new solver is demonstrated by comparisons with a conventional line Gauss-Siedel Relaxation solver. Computational test results for transonic flow (trailing edge flow in a transonic turbine cascade) and hypersonic flow (M = 6.0 shock-on-shock phenoena on a cylindrical leading edge) are presented. When applied to the transonic cascade case, the new solver is 4.4 times faster in terms of number of iterations and 3.1 times faster in terms of CPU time than the Relaxation solver. For the hypersonic shock case, the new solver is 3.0 times faster in terms of number of iterations and 2.2 times faster in terms of CPU time than the Relaxation solver.

Ajmani, Kumud↗

Advanced propeller research

Recent results of aerodynamic and acoustic research on both single rotation and counterrotation propellers are reviewed. Data and analytical results are presented for three propellers: SR-7A, the single rotation design used in the NASA Propfan Test Assessment (PTA) flight program; CRP-X1, the initial 5+5 Hamilton Standard counterrotating design; and F7-A7, the 8+8 counterrotating G.E. design used in the proof of concept Unducted Fan (UDF) engine. In addition to propeller efficiencies, cruise and takeoff noise, and blade pressure data, off-design phenomena involving formation of leading edge vortexes are described. Aerodynamic and acoustic computational results derived from 3-D Euler and acoustic radiation codes are presented. Research on unsteady flows which are particularly important for understanding counterrotation interaction noise, unsteady loading effects on acoustics, and flutter or forced response is described. The first results of 3-D unsteady Euler solutions are illustrated for a single rotation propeller at angle of attack and for a counterrotation propeller. Basic experimental and theoretical results from studies on the unsteady aerodynamics of oscillating cascades are outlined.

Groeneweg, John F.↗

Postbuckling of long thick plates in compression including higher order transverse shearing effects

Aluminum plates of various length/width ratios loaded in compression are examined analytically in terms of buckling and postbuckling for comparison with classical theory. The plates are considered to be subjected to longitudinal compressive displacement with no edge stress and are assumed to be simply supported at the side edges. The average longitudinal direct stresses are computed to plot characteristic curves by means of the following theories: 3D flexibility, higher-order shear deformation, first-order shear deformation, and classical von Karman. The 3D flexibility approach yields the lowest results with more unknowns than the higher-order shear-deformation method. The 3D flexibility approach is considered to be the most accurate, and it is shown that the calculated resultants and displacements tend to vary when consideration is given to the effects of transverse shear.

Stein, Manuel↗