Augmented Physics-Informed Neural Networks (APINNs): A gating network-based soft domain decomposition methodology
Not provided.
Automation & Control Systems↗
SEARCH · Search NASA
Results for “AUGMENTATION”
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.
Machine Learning Augmented Predictive and Generative Model for Rupture Life in Ferritic and Austenitic Steels
The Larson-Miller parameter (LMP) offers an efficient and fast scheme to estimate the creep rupture life of alloy materials for high temperature applications. However, owing to poor generalizability and dependence on the constant C, which is typically not known a-priori, estimations using the Larson-Miller parameter often result in suboptimal performance for a wide range of materials. At best it is useful in comparing alloys of similar composition. In this work, three machine learning (ML) schemes were developed for rupture life prediction for 9-12% Cr ferritic-martensitic steels and austenitic stainless steels, i.e., a hierarchical model to parameterize LMP using the LMP constant C to compute rupture life, a hierarchical model to parameterize both C and LMP to compute rupture life, and a direct prediction of rupture life. Specifically, we show that the third scheme, using a gradient boosting algorithm, can be used to train ML models for very accurate prediction of rupture life in a variety of alloys (Pear-son Correlation Coefficient > 0.9 for 9-12% Cr and > 0.8 for austenitic stainless steels). In addition, the Shapley value was used to quantify feature importance, making the model interpretable by identifying the effect of various features on the model performance. Furthermore, a variational autoencoder-based generative model was built by conditioning on the experimental dataset to sample hypothetical synthetic candidate alloys from the learnt joint distribution not existing in both 9-12 % Cr ferritic-martensitic steel and austenitic stainless steel datasets. Finally, based on the predictive and generative model, a reinforcement learning strategy has been proposed to guide experimentalists into designing better heat resistant alloys.
Mamun, Md Osman G.↗
Flexspander: augmenting expander networks in high-performance systems with optical bandwidth steering
Communication efficiency is one of the deciding factors in determining many of today’s high-performance computing (HPC) applications. Traditionally, HPC systems have been on static network topologies, making them inflexible to the variety of skewed traffic demands that may arise due to the spatial locality inherent in many applications. To handle traffic locality, researchers have proposed integrating optical circuit switches (OCSs) into the network architecture, which reconfigures the network topology to alter and dynamically adapt to the predicted traffic. In this paper, we present a novel reconfigurable network topology called Flexspander. Beyond offering a flexible interconnect, Flexspander also offers full flexibility in terms of construction and can be built with any arbitrary combination of commercial electrical packet switches and OCSs. We evaluate Flexspander performance through extensive simulations with multiple network traces, and our results show improved performance for Flexspander over currently proposed static and reconfigurable topologies in terms of the flow completion time.
Teh, Min Yee↗
Enhancing Electron Microscopy Image Classification Using Data Augmentation
Manual labeling for machine learning tasks such as image classification is tedious and labor-intensive; as a result, scientific datasets suitable for deep learning applications are scarce and limited. While data augmentation techniques have shown promise for extending image datasets, very little work has been done to understand the impact of combining multiple augmentation methods sequentially or the limits of their effectiveness when combined. Our work addresses this gap by examining how standard and combinatorial data augmentation affects the performance of machine learning models when trained on small datasets for label classification tasks. For our analysis, we generate single, double and quadruple-augmented datasets for a microscopy image classification task using six standard augmentation methods, and compare the resultant improvements observed in binary classification accuracy with three standard image classification models (DenseNet169, MobileNetV2, ResNet101V2). Our experiments show a non-monotonic relationship between the number of simultaneous augmentation methods and classification accuracy, indicating that there is a trade-off between the degree of augmentation and the model performance. These findings suggest that the optimal number of augmentation methods will vary by domain and use case. We also find that the order in which augmentation methods are applied to a limited dataset matters when combining augmentation schemes, with our use case showing performance differences up to 2.6% when the augmentation order is reversed for double-augmented datasets. Our work offers insights to the limits of data augmentation when working on image classification tasks with limited datasets.
Welsman, Jordan A↗
Percussive Augmenter of Rotary Drills (PARoD)
A percussive augmenter is being developed and it is designed to add ultrasonic hammering to a conventional commercial rotary drill through an adapter using a piezoelectric actuator. The combination of ultrasonic hammering and rotation creates a drill that requires low power and low axial load while providing faster penetration than conventional rotary-only drills. These characteristics make percussive augmentation of rotary drills ideal for not only planetary exploration but commercial applications as well. The purpose of this internship was to produce, test, and optimize an augmenter that drives a 2 inch diameter bit. This task was part of the percussive augmenter's phase II of an ongoing contract and it involved increasing the capability of the .25 inch version of the augmenter prototype that was developed in phase I. Due to fabrication delays of the augmenter, an extensive part of the test effort was conducted using previous rotary and hammer drill prototypes. These tests focused on drilling deep over long periods of time to provide for effective stress test of the drill. Modifications were made to the drill, its components, and the testing procedure to achieve desired borehole depths. These results were used to start initial testing on the 2 inch augmenter and provide preliminary data on the augmenter's ability to significantly improve penetration rate over conventional rotary-only drills.
rotraty hammer↗
Data augmentation and multimodal learning for predicting drug response in patient-derived xenografts from gene expressions and histology images
Patient-derived xenografts (PDXs) are an appealing platform for preclinical drug studies. A primary challenge in modeling drug response prediction (DRP) with PDXs and neural networks (NNs) is the limited number of drug response samples. We investigate multimodal neural network (MM-Net) and data augmentation for DRP in PDXs. The MM-Net learns to predict response using drug descriptors, gene expressions (GE), and histology whole-slide images (WSIs). We explore whether combining WSIs with GE improves predictions as compared with models that use GE alone. We propose two data augmentation methods which allow us training multimodal and unimodal NNs without changing architectures with a single larger dataset: 1) combine single-drug and drug-pair treatments by homogenizing drug representations, and 2) augment drug-pairs which doubles the sample size of all drug-pair samples. Unimodal NNs which use GE are compared to assess the contribution of data augmentation. The NN that uses the original and the augmented drug-pair treatments as well as single-drug treatments outperforms NNs that ignore either the augmented drug-pairs or the single-drug treatments. In assessing the multimodal learning based on the MCC metric, MM-Net outperforms all the baselines. Our results show that data augmentation and integration of histology images with GE can improve prediction performance of drug response in PDXs.
60 APPLIED LIFE SCIENCES↗
Thrust Augmentation of a Turbojet Engine at Simulated Flight Conditions by Introduction of a Water-Alcohol Mixture into the Compressor
An investigation was conducted at simulated high-altitude flight conditions to evaluate the use of compressor evaporative cooling as a means of turbojet-engine thrust augmentation. Comparison of the performance of the engine with water-alcohol injection at the compressor inlet, at the sixth stage of the compressor, and at the sixth and ninth stages was made. From consideration of the thrust increases achieved, the interstage injection of the coolant was considered more desirable preferred over the combined sixth- and ninth-stage injection because of its relative simplicity. A maximum augmented net-thrust ratio of 1.106 and a maximum augmented jet-thrust ratio of 1.062 were obtained at an augmented liquid ratio of 2.98 and an engine-inlet temperature of 80 F. At lower inlet temperatures (-40 to 40 F), the maximum augmented net-thrust ratios ranged from 1.040 to 1.076 and the maximum augmented jet-thrust ratios ranged from 1.027 to 1.048, depending upon the inlet temperature. The relatively small increase in performance at the lower inlet-air temperatures can be partially attributed to the inadequate evaporation of the water-alcohol mixture, but the more significant limitation was believed to be caused by the negative influence of the liquid coolant on engine- component performance. In general, it is concluded that the effectiveness of the injection of a coolant into the compressor as a means of thrust augmentation is considerably influenced by the design characteristics of the components of the engine being used.
Useller, James W.↗
Studies of Operating Frequency Effects On Ejector-based Thrust Augmentation in a Pulse Detonation Engine
Studies were performed in order to characterize the thrust augmentation potential of an ejector in a Pulse Detonation Engine application. A 49-mm diameter tube of 0.914-m length was constructed with one open end and one closed end. Ethylene, oxygen, and nitrogen were introduced into the tube at the closed end through the implementation of a fast mixing injector. The tube was completely filled with a stoichiometric mixture containing a one to one molar ratio of nitrogen to oxygen. Ethylene was selected as the fuel due to its detonation sensitivity and the molar ratio of the oxidizer was chosen for heat transfer purposes. Detonations were initiated in the tube through the use of a spark ignition system. The PDE was operated in a multi-cycle mode at frequencies ranging from 20-Hz to 50-Hz. Baseline thrust measurements with no ejector present were performed while operating the engine at various frequencies and compared to theoretical estimates. The baseline values were observed to agree with the theoretical model at low operating frequencies and proved to be increasingly lower than the predicted values as the operating frequency was increased. The baseline thrust measurements were observed to agree within 15 percent of the model for all operating frequencies. A straight 152-mm diameter ejector was installed and thrust augmentation percentages were measured. The length of the ejector was varied while the overlap percentage (percent of the ejector length which overlapped the tube) was maintained at 25 percent for all tests. In addition, the effect of ejector inlet geometry was investigated by comparing results with a straight inlet to those of a 38-mm inlet diameter. The thrust augmentation of the straight inlet ejector proved to be independent of engine operating frequency, augmenting thrust by 40 percent for the 0.914-m length ejector. In contrast, the rounded lip ejector of the same length seemed to be highly dependent on the engine operating frequency. An optimum operating frequency observed with the rounded inlet occurred at an operating frequency of 30-Hz, resulting in thrust augmentation percentages greater than 100 percent. The effect that the engine operating frequency had on thrust augmentation levels attained with an ejector was characterized and optimum performance parameters were established. Insight into the frequency dependent nature of the ejector performance was pursued. Suggestions for future experiments which are needed to fully understand the means in which thrust augmentation is achieved in a PDE-ejector configuration were noted.
Landry, K.↗
Making Invisible Visible: Data-Driven Seismic Inversion With Spatio-Temporally Constrained Data Augmentation
Deep learning and data-driven approaches have shown great potential in scientific domains. The promise of data-driven techniques relies on the availability of a large volume of high-quality training datasets. Due to the high cost of obtaining data through expensive physical experiments, instruments, and simulations, data augmentation techniques for scientific applications have emerged as a new direction for obtaining scientific data recently. However, existing data augmentation techniques originating from computer vision yield physically unacceptable data samples that are not helpful for the domain problems that we are interested in. In this article, we develop new data augmentation techniques based on convolutional neural networks. Specifically, our generative models leverage different physics knowledge (such as governing equations, observable perception, and physics phenomena) to improve the quality of the synthetic data. To validate the effectiveness of our data augmentation techniques, we apply them to solve a subsurface seismic full-waveform inversion using simulated CO 2 leakage data. Our interest is to invert for subsurface velocity models associated with very small CO 2 leakage. We validate the performance of our methods using comprehensive numerical tests. Here via comparison and analysis, we show that data-driven seismic imaging can be significantly enhanced by using our data augmentation techniques. Particularly, the imaging quality has been improved by 15% in test scenarios of general-sized leakage and 17% in small-sized leakage when using an augmented training set obtained with our techniques.
58 GEOSCIENCES↗
Orion MPCV Continuum RCS Heating Augmentation Model Development
The reaction control system jets of the Orion Multi Purpose Crew Vehicle can have a significant impact on the magnitude and distribution of the surface heat flux on the leeside of the aft-body, when they are fired. Changes in surface heating are expressed in terms of augmentation factor over the baseline smooth body heating. Wind tunnel tests revealed heating augmentation factors as high as 13.0, 7.6, 2.8, and 5.8 for the roll, pitch down, pitch up, and yaw jets respectively. Heating augmentation factor models, based almost exclusively on data from a series of wind tunnel tests have been developed, for the purposes of thermal protection system design. The wind tunnel tests investigated several potential jet-to-freestream similarity parameters, and heating augmentation factors derived from the data showed correlation with the jet-to-freestream momentum ratio. However, this correlation was not utilized in the developed models. Instead augmentation factors were held constant throughout the potential trajectory space. This simplification was driven by the fact that ground to flight traceability and sting effects are not well understood. Given the sensitivity of the reaction control system jet heating augmentation to configuration, geometry, and orientation the focus in the present paper is on the methodology used to develop the models and the lessons learned from the data. The models that are outlined in the present work are specific to the aerothermal database used to design the thermal protection system for the Exploration Flight Test 1 vehicle.
Heating Augmentation↗
Molecular phylogenetic trees - On the validity of the Goodman-Moore augmentation algorithm
A response is made to the reply of Nei and Tateno (1979) to the letter of Holmquist (1978) supporting the validity of the augmentation algorithm of Moore (1977) in reconstructions of nucleotide substitutions by means of the maximum parsimony principle. It is argued that the overestimation of the augmented numbers of nucleotide substitutions (augmented distances) found by Tateno and Nei (1978) is due to an unrepresentative data sample and that it is only necessary that evolution be stochastically uniform in different regions of the phylogenetic network for the augmentation method to be useful. The importance of the average value of the true distance over all links is explained, and the relative variances of the true and augmented distances are calculated to be almost identical. The effects of topological changes in the phylogenetic tree on the augmented distance and the question of the correctness of ancestral sequences inferred by the method of parsimony are also clarified.
Holmquist, R.↗
Cooperative synthesis of control and display augmentation for a STOL aircraft in the approach and landing task
Application of the Cooperative methodology to synthesize control/display augmentation systems for the piloted longitudinal landing task with a modern, statically-unstable, fighter aircraft is considered. Starting with a control augmentation law which yields augmented vehicle dynamics that meet Level I handling qualities specifications, the effect of time-delay in the Head-up-display is studied using model-based criterion. This evaluation showed that even with 'good' conventional dynamics, a realistic value of display time-delay will cause significant deterioration in pilot workload and piloted-system performance. Application of the Cooperative methodology to control augmentation alone resulted in augmented vehicle dynamics which provide direct control of the flight path from the pilot stick. Analytical evaluation of these dynamics indicates that such dynamics might lead to improved pilot ratings over conventional dynamics, especially in the presence of time-delays in the system. Also, application of the methodology to simultaneous synthesis of control augmentation and flight director laws revealed that it might be advantageous to consider the control/display trade-off in the early design stages.
Garg, Sanjay↗
Low-Speed Wind-Tunnel Investigation to Determine the Aerodynamic Characteristics of a Rectangular Wing Equipped with a Full-Span and an Inboard Half-Span Jet-Augmented Flap Deflected 55 deg
An investigation to determine the aerodynamic characteristics of a rectangular wing equipped with a full-span and an inboard half-span jet-augmented flap has been made in the Langley 300 MPH 7- by 10-foot tunnel. The wing had an aspect ratio of 8.3 and a thickness-chord ratio of 0.167. A jet of air was blown backward through a small gap, tangentially to the upper surface of a round trailing edge, and was separated from the trailing edge by a very small flap at an angle of 55 deg with respect to the wing-chord plane. The results of the investigation showed that the ratio of total lift to jet-reaction lift for the wing was about 35 percent less for the half-span jet-augmented flap than for the full-span jet-augmented flap. The reduction of the span of the jet-augmented flap from full to half span reduced the maximum value of jet-circulation lift coefficient that could be produced from about 6.8 to a value of about 2.2. The half-span jet- augmented flap gave thrust recoveries considerably poorer than those obtained with the full-span jet-augmented flap. Large nose-down pitching- moment coefficients were produced by the half-span flap, with the greater part of these being the result of the larger jet reactions required to produce a given lift for the half-spin flap compared with that required for the full-span flap.
Gainer, Thomas G.↗
Parametric Investigation of Thrust Augmentation by Ejectors on a Pulsed Detonation Tube
A parametric investigation has been made of thrust augmentation of a 1 inch diameter pulsed detonation tube by ejectors. A set of ejectors was used which permitted variation of the ejector length, diameter, and nose radius, according to a statistical design of experiment scheme. The maximum augmentations for each ejector were fitted using a polynomial response surface, from which the optimum ejector diameters, and nose radius, were found. Thrust augmentations above a factor of 2 were measured. In these tests, the pulsed detonation device was run on approximately stoichiometric air-hydrogen mixtures, at a frequency of 20 Hz. Later measurements at a frequency of 40 Hz gave lower values of thrust augmentation. Measurements of thrust augmentation as a function of ejector entrance to detonation tube exit distance showed two maxima, one with the ejector entrance upstream, and one downstream, of the detonation tube exit. A thrust augmentation of 2.5 was observed using a tapered ejector.
Wilson, Jack↗
Parametric Investigation of Thrust Augmentation by Ejectors on a Pulsed Detonation Tube
A parametric investigation has been made of thrust augmentation of a 1 in. diameter pulsed detonation tube by ejectors. A set of ejectors was used which permitted variation of the ejector length, diameter, and nose radius, according to a statistical design of experiment scheme. The maximum augmentation ratios for each ejector were fitted using a polynomial response surface, from which the optimum ratios of ejector diameter to detonation tube diameter, and ejector length and nose radius to ejector diameter, were found. Thrust augmentation ratios above a factor of 2 were measured. In these tests, the pulsed detonation device was run on approximately stoichiometric air-hydrogen mixtures, at a frequency of 20 Hz. Later measurements at a frequency of 40 Hz gave lower values of thrust augmentation. Measurements of thrust augmentation as a function of ejector entrance to detonation tube exit distance showed two maxima, one with the ejector entrance upstream, and one downstream, of the detonation tube exit. A thrust augmentation of 2.5 was observed using a tapered ejector.
Wilson, Jack↗
Performance Evaluation of Augmented Teleoperation of Contact Manipulation Tasks
At Argonne National Laboratory, an enhanced telerobotic concept has been suggested by providing augmented virtual fixtures for precise and dexterous manipulation. The proposed technology incorporated a human-robot interface that can assist personnel on the processes of Deactivation and Decommissioning in sites of high radiation. It has been implemented on a dual-arm robot system, known as Baxter, by integrating the Robotic Operating System (ROS) with 3D sensors and visual-haptic operator interface. However, what remained before its practical utilization is the capability for precisely placing the virtual fixtures. To this end, the objectives involved camera calibration, which required hardware fabrication and setup, as well as modification and implementation of ROS. The goal of this Research and Development is to develop an augmented-reality- based operator interface for human-inclusive, intelligent, and autonomous robotic system for the Deactivation and Decommissioning (D and D) of radioactive sites. The objectives are to develop: Collaborative Robot; Multi-modal Mixed-Reality Human-Robot Interface; Intelligent and Autonomous Behaviors. In order to achieve this, Baxter's Head-Kinect camera and Left-Hand camera need to be calibrated to overlap the real environment with the simulated environment. Camera calibration, which is crucial for precisely placing the virtual fixtures, was successful after running the simulations. The real and simulated environments overlapped, permitting the utilization of augmented teleoperation and the dexterous manipulation of Baxter for D and D applications in nuclear sites. Future steps: Performance evaluation of augmented teleoperation of contact manipulation tasks: Multiple-Camera 3D Reconstruction and Automation. Oculus Interface: Augmented reality and machine learning for enhanced telerobotic operation (Augmented Tele-autonomy): Machine-learning techniques for complex and flexible autonomous robotic behaviors. Interactive remote operation of robotic systems. Mesh cutting with simulation and real robot.
12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
An adaptive sampling augmented Lagrangian method for stochastic optimization with deterministic constraints
The primary goal of this paper is to provide an efficient solution algorithm based on the augmented Lagrangian framework for optimization problems with a stochastic objective function and deterministic constraints. Our main contribution is combining the augmented Lagrangian framework with adaptive sampling, resulting in an efficient optimization methodology validated with practical examples. To achieve the presented efficiency, here we consider inexact solutions for the augmented Lagrangian subproblems, and through an adaptive sampling mechanism, we control the variance in the gradient estimates. Furthermore, we analyze the theoretical performance of the proposed scheme by showing equivalence to a gradient descent algorithm on a Moreau envelope function, and we prove sublinear convergence for convex objectives and linear convergence for strongly convex objectives with affine equality constraints. The worst-case sample complexity of the resulting algorithm, for an arbitrary choice of penalty parameter in the augmented Lagrangian function, is $\mathscr{O}$(ϵ -3-δ ) , where ϵ > 0 is the expected error of the solution and δ > 0 is a user-defined parameter. If the penalty parameter is chosen to be $\mathscr{O}$(ϵ -1 ), we demonstrate that the result can be improved to $\mathscr{O}$(ϵ -2 ) , which is competitive with the other methods employed in the literature. Moreover, if the objective function is strongly convex with affine equality constraints, we obtain $\mathscr{O}$(ϵ -1 log(1/ϵ)) complexity. Finally, we empirically verify the performance of our adaptive sampling augmented Lagrangian framework in machine learning optimization and engineering design problems, including topology optimization of a heat sink with environmental uncertainty.
97 MATHEMATICS AND COMPUTING↗