SEARCH · Engineering Papers
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.
Deep Space Network - tracking and navigational accuracy analysis, communications research, development, and engineering, and Deep Space Station engineering and operations
DINET is a technology development experiment intended to increase the technical readiness of JPL s implementation of DTN protocols - "ION". The objective is to advance ION in flight and ground SW to TRL 8, with code of sufficient quality that future flight projects can easily use it at low risk. DINET is to be implemented on the Deep Impact flyby spacecraft. DINET operations will be performed in late 2008 during the Deep Impact spacecraft team "stand down" after EPOCH operations and before the start of development for DIXI operations.
In order to promote preclinical and clinical applications of photoacoustic imaging, novel photoacoustic contrast agents are highly desired for molecular imaging of diseases, especially for deep tumor imaging. In this paper, perylene-3,4,9,10-tetracarboxylic diiimide-based near-infrared-absorptive organic nanoparticles are reported as an efficient agent for photoacoustic imaging of deep brain tumors in living mice with enhanced permeability and retention effect
There is an increasing demand for the development of ion-conducting electrolytes for energy storage systems. Much attention is directed toward deep eutectic solvents as potential candidates. In the search for highly conductive systems, the possibility of designing deep eutectic solvents with Grotthuss-type proton transport is widely overlooked. Herein, ethaline, a mixture of choline chloride and ethylene glycol is used in a 1:2 molar ratio, to induce a significant conductivity increase with the addition of water and sulfuric acid (H 2 SO 4 ). The achieved breakthrough conductivity is analyzed experimentally and simulated with ab initio molecular dynamics (AIMD). At sufficient water content, an H-bonding network is formed that leads to a significant breakthrough conductivity based on H 2 SO 4 -derived proton transfer following the long-established Grotthuss proton transport mechanism. This result is substantiated by the positive deviation from the ideal KCl line in the Walden plot. Specifically, the data series positioned above the reference line indicates a Grotthuss mechanism in action. The AIMD simulations demonstrate proton transfer between water and ethylene glycol, supported by simulation frames captured at various times.
Solution-processed quasi-2D perovskites contain multiple quantum wells with a broad width distribution. Inhomogeneity results in the charge funneling into the smallest bandgap components, which hinders deep-blue emission and accelerates Auger recombination. Here, a synthetic strategy applied to a range of quasi-2D perovskite systems is reported, that significantly narrows the quantum well dispersity. It is shown that the phase distribution in the perovskite film is significantly narrowed with controlled, simultaneous evaporation of solvent and antisolvent. Modulation of film formation kinetics of quasi-2D perovskite enables stable deep-blue electroluminescence with a peak emission wavelength of 466 nm and a narrow linewidth of 14 nm. Light emitting diodes using the perovskite film show a maximum luminance of 280 cd m –2 at an external quantum efficiency of 0.1%. This synthetic approach will serve in producing new materials widening the color gamut of next-generation displays.
The functionality of ferroelastic domain walls in ferroelectric materials is explored in real-time via the in situ implementation of computer vision algorithms in scanning probe microscopy (SPM) experiment. The robust deep convolutional neural network (DCNN) is implemented based on a deep residual learning framework (Res) and holistically nested edge detection (Hed), and ensembled to minimize the out-of-distribution drift effects. The DCNN is implemented for real-time operations on SPM, converting the data stream into the semantically segmented image of domain walls and the corresponding uncertainty. Further the pre-defined experimental workflows perform piezoresponse spectroscopy measurement on thus discovered domain walls, and alternating high- and low-polarization dynamic (out-of-plane) ferroelastic domain walls in a PbTiO 3 (PTO) thin film and high polarization dynamic (out-of-plane) at short ferroelastic walls (compared with long ferroelastic walls) in a lead zirconate titanate (PZT) thin film is reported. This work establishes the framework for real-time DCNN analysis of data streams in scanning probe and other microscopies and highlights the role of out-of-distribution effects and strategies to ameliorate them in real time analytics.
Deep neural network (DNN) surrogates of expensive physics simulations are enabling a rapid change in the way that common experimental design and analysis tasks are approached. Surrogate models allow simulations to be performed in parallel and separately from downstream tasks, thereby enabling analyses that would be impossible with the simulation in-the-loop; surrogates based on DNNs can effectively emulate diverse non-scalar data of the types collected in fusion and laboratory-astrophysics experiments. The challenge is in training the surrogate model, for which large ensembles of physics simulations must be run, preferably without wasting computational effort on uninteresting simulations. Here, in this paper, we present an iterative sampling scheme that can preferentially propose simulations in interesting regions of parameter space without neglecting unexplored regions, allowing high-quality and wide-ranging surrogate models to be trained using 2–3 times fewer simulations compare to space-filling designs. Our approach uses an explicit importance function defined on the simulation output space, balanced against a measure of simulation density which serves as a proxy for surrogate accuracy. It is easy to implement and can be tuned to find interesting simulations early in the study, allowing surrogates to be trained quickly and refined as new simulations become available; this represents an important step towards the routine generation of deep surrogate models quickly enough to be truly relevant to experimental work.
A novel and modern framework for energy modeling is developed in this paper with a focus on nuclear energy modeling and simulation. The framework combines multiphysics simulations and real data, with validation by uncertainty quantification tasks and facilitation by machine and deep learning methods. The hybrid framework is built on the basis of a wide range of physical models, real data, mathematical and statistical methods, and artificial intelligence techniques. The framework is demonstrated in different applications, including quantifying uncertainties in computer simulations, multiphysics coupling, analysis of variance using machine learning surrogate models, deep learning of time series phenomena, and propagating parametric uncertainties of nuclear data. The applications demonstrated are oriented to nuclear engineering simulations, even though majority of the methods are applicable to other energy sources (eg, renewable). Efficient utilization of this framework is expected to yield a much better understanding of the physical phenomena analyzed as well as an improvement in the performance of the energy design/model under construction.
The formation of first raindrops in deep convective clouds is investigated. A combination of observational data analysis and 2-D and 3-D numerical bin microphysical simulations of deep convective clouds suggests that the first raindrops form at the top of undiluted or slightly diluted cores. It is shown that droplet size distributions in these regions are wider and contain more large droplets than in diluted volumes. The results of the study indicate that the initial raindrop formation is determined by the basic microphysical processes within ascending adiabatic volumes. It allows one to predict the height of the formation of first raindrops considering the processes of nucleation, diffusion growth and collisions. The results obtained in the study explain observational results reported by Freud and Rosenfeld (2012) according to which the height of first raindrop formation depends linearly on the droplet number concentration at cloud base. The results also explain why a simple adiabatic parcel model can reproduce this dependence. The present study provides a physical basis for retrieval algorithms of cloud microphysical properties and aerosol properties using satellites proposed by Rosenfeld et al. ( 2012). The study indicates that the role of mixing and entrainment in the formation of the first raindrops is not of crucial importance. It is also shown that low variability of effective and mean volume radii along horizontal traverses, as regularly observed by in situ measurements, can be simulated by high-resolution cloud models, in which mixing is parameterized by a traditional 1.5 order turbulence closure scheme.
We present an extraction of unpolarized transverse-momentum-dependent parton distribution and fragmentation functions based on more than two thousand data points from several experiments for two different processes: semi-inclusive deep-inelastic scattering and Drell-Yan production. The baseline analysis is performed using the Monte Carlo replica method and resumming large logarithms at N 3 LL accuracy. The resulting description of the data is very good (χ 2 /N dat = 1.06). For semi-inclusive deep-inelastic scattering, predictions for multiplicities are normalized by factors that cure the discrepancy with data introduced by higher-order perturbative corrections.
This paper presents high precision control and deep learning-based corn stand counting algorithms for a low-cost, ultra-compact 3D printed and autonomous field robot for agricultural operations. Currently, plant traits, such as emergence rate, biomass, vigor, and stand counting, are measured manually. This is highly labor-intensive and prone to errors. The robot, termed TerraSentia, is designed to automate the measurement of plant traits for efficient phenotyping as an alternative to manual measurements. In this paper, we formulate a Nonlinear Moving Horizon Estimator that identifies key terrain parameters using onboard robot sensors and a learning-based Nonlinear Model Predictive Control that ensures high precision path tracking in the presence of unknown wheel-terrain interaction. Moreover, we develop a machine vision algorithm designed to enable an ultra-compact ground robot to count corn stands by driving through the fields autonomously. The algorithm leverages a deep network to detect corn plants in images, and a visual tracking model to re-identify detected objects at different time steps. We collected data from 53 corn plots in various fields for corn plants around 14 days after emergence (stage V3 - V4). The robot predictions have agreed well with the ground truth with C robot =1.02×C human -0.86 and a correlation coefficient R=0.96. The mean relative error given by the algorithm is -3.78%, and the standard deviation is 6.76%. These results indicate a first and significant step towards autonomous robot-based real-time phenotyping using low-cost, ultra-compact ground robots for corn and potentially other crops.
Tropical regions hold one third of the world’s soil organic carbon, but few experiments have warmed tropical soils in situ. The vulnerability of these soils to climate change-induced losses is uncertain with many hypothesizing these soils would be less sensitive to climate change because already-high temperatures in tropical systems might limit microbial sensitivity or due to increased mineral protection of organic carbon in highly weathered tropical soils. Here we present the results of a deep soil (0–100 cm) warming experiment in a tropical Andisol. Andisols can store large, persistent pools of soil carbon that are protected from decomposition by poorly and non-crystalline minerals (PNCM). In 20 cm depth intervals, we measured key soil properties including carbon, nitrogen, pH, PNCM, bacterial and fungal richness along with temperature, moisture, and CO 2 production. Over a year of soil warming, CO 2 production significantly increased by 50–300% per degree of warming, but only in the top 40 cm of the soil profile in contrast to the results of other deep soil warming experiments. Multimodal analysis supported our hypothesis that high concentrations of PNCM was the primary driver of the lack of CO 2 response, followed by high relative soil moisture and low bacterial richness, which may be a proxy for organic carbon availability. The lack of elevated CO 2 production in response to warming suggests a limited positive feedback to climate change in Andisols driven by their strong mineral protection of organic matter. Therefore, Andisols should be considered high priority restoration or protection areas when considering the management of soil carbon stocks as part of climate action.
Deep eutectic solvent (DES) pretreatment coupled with alkaline degumming on Apocynum venetum bast was studied for developing an eco-friendly and effective degumming method. Here, a 1:2 mixture of choline chloride and urea was used in a microwave-assisted DES pretreatment optimized to 60 min and 110 °C. The impact of the deep eutectic solvent pretreatment on the properties of produced semi-degummed and refined dry fibers were assessed by examining the fiber chemical composition, surface microstructure, cellulose crystallinity, degree of polymerization, and fiber thermal properties. The combined DES and alkaline degumming treatment effectively removed the gummy matter including hemicellulose and lignin from the A. venetum bast, providing a smooth and clean fiber surface. This combined DES and alkaline degumming treatment also shows the ability to extract amorphous cellulose, leading to an increase of fiber crystallinity and a decrease of the average cellulose molecular weight. The thermal stability of the fibers was improved significantly after the degumming treatment. The refined dry fibers produced by this combined DES-alkaline treatment exhibited comparable chemical and physical properties (e.g. residual gum content, fiber length, fiber fineness, breaking tenacity) compared with traditionally alkali degumming method. These results revealed that DES is a practical and feasible pretreatment protocol for A. venetum bast degumming.
Here this article proposes a model-based deep reinforcement learning (DRL) method to design emergency control strategies for short-term voltage stability problems in power systems. Recent advances show promising results for model-free DRL-based methods in power systems control problems. But in power systems applications, these model-free methods have certain issues related to training time (clock time) and sample efficiency; both are critical for making state-of-the-art DRL algorithms practically applicable. DRL-agent learns an optimal policy via a trial-and-error method while interacting with the real-world environment. It is also desirable to minimize the direct interaction of the DRL agent with the real-world power grid due to its safety-critical nature. Additionally, the state-of-the-art DRL-based policies are mostly trained using a physics-based grid simulator where dynamic simulation is computationally intensive, lowering the training efficiency. We propose a novel model-based DRL framework where a deep neural network (DNN)-based dynamic surrogate model (SM), instead of a real-world power grid or physics-based simulation, is utilized within the policy learning framework, making the process faster and more sample efficient. However, having stable training in model-based DRL is challenging because of the complex system dynamics of large-scale power systems. We addressed these issues by incorporating imitation learning to have a warm start in policy learning, reward-shaping, and multi-step loss in surrogate model training. Finally, we achieved 97.5% reduction in samples and 87.7% reduction in training time for an application to the IEEE 300-bus test system.
Austenitic 347H stainless steel offers superior mechanical properties and corrosion resistance required for extreme operating conditions such as high temperature. The change in microstructure due to composition and process variations is expected to impact material properties. Identifying microstructural features such as grain boundaries thus becomes an important task in the process-microstructure-properties loop. Applying convolutional neural network (CNN)-based deep learning models is a powerful technique to detect features from material micrographs in an automated manner. In contrast to microstructural classification, supervised CNN models for segmentation tasks require pixel-wise annotation labels. However, manual labeling of the images for the segmentation task poses a major bottleneck for generating training data and labels in a reliable and reproducible way within a reasonable timeframe. Microstructural characterization especially needs to be expedited for faster material discovery by changing alloy compositions. Here, in this study, we attempt to overcome such limitations by utilizing multimodal microscopy to generate labels directly instead of manual labeling. We combine scanning electron microscopy images of 347H stainless steel as training data and electron backscatter diffraction micrographs as pixel-wise labels for grain boundary detection as a semantic segmentation task. The viability of our method is evaluated by considering a set of deep CNN architectures. We demonstrate that despite producing instrumentation drift during data collection between two modes of microscopy, this method performs comparably to similar segmentation tasks that used manual labeling. Additionally, we find that naïve pixel-wise segmentation results in small gaps and missing boundaries in the predicted grain boundary map. By incorporating topological information during model training, the connectivity of the grain boundary network and segmentation performance is improved. Finally, our approach is validated by accurate computation on downstream tasks of predicting the underlying grain morphology distributions which are the ultimate quantities of interest for microstructural characterization.
X-ray diffraction patterns contain information about the atomistic structure and microstructure (defect population) of materials, extracting detailed information from diffraction patterns is complex, demanding and relies on prior knowledge. Here, we hypothesize that deep-learning techniques can help to perform an effective and accurate analysis with high throughput rates. To demonstrate this concept, we applied a novel deep learning framework to determine the evolution of the β-phase volume fraction in a Ti–6Al–4V alloy during heat-treatment from video sequences of 2D diffraction patterns recorded in transmission and with highly monochromatic radiation in a synchrotron beamline. In particular, we studied the impact of network design on prediction reliability and computational performance. Networks of different architectures were trained using 3008 experimental 2D patterns. A well-tuned model was found to reproduce the phase fractions of another experimental data set, consisting of 1100 diffraction patterns, with a mean-square error as small as 2.6 x 10 -4 . The average prediction error of β-phase volume fraction was within 1.6 x 10 -2 (in each diffraction pattern) of the values obtained by conventional methods. Our work demonstrates that convolutional neural networks can evaluate high energy X-ray diffraction patterns with a remarkable level of reliability. Furthermore, it demonstrates the significance of network design on the reliability of predictions and computational performance. The most complex models do not necessarily result in highest accuracy and may even fail to learn from the data.
Deep generative models (DGMs) can generate synthetic data samples that closely resemble the original dataset, addressing data scarcity. In this work, we developed a conditional variational autoencoder (CVAE) to augment critical heat flux (CHF) data used for the 2006 Groeneveld lookup table. To compare with traditional methods, a fine-tuned deep neural network (DNN) regression model was evaluated on the same dataset. Both models achieved small mean absolute relative errors, with the CVAE showing more favorable results. Uncertainty quantification (UQ) was performed using repeated CVAE sampling and DNN ensembling. The DNN ensemble improved performance over the baseline, while the CVAE maintained consistent results with less variability and higher confidence. Both models achieved small errors inside and outside the training domain, with slightly larger errors outside. Altogether, the CVAE performed better than the DNN in predicting CHF and exhibited better uncertainty behavior.
Accurate knowledge of the distribution system topology and parameters is required to achieve good voltage control performance, but this is difficult to obtain in practice. This paper proposes a physical-model-free voltage control method based on a surrogate-model-enabled deep reinforcement learning approach. Specifically, a surrogate model is trained in a supervised manner using the recorded limited number of historical data to learn the relationship between the power injections and voltage fluctuations of each node. Then, the deep reinforcement learning algorithm is applied to learn an optimal control strategy from the experiences obtained by continuous interactions with the surrogate model. The proposed method can achieve physical-model-free control of unbalanced distribution network and inform real-time decisions to deal with fast voltage fluctuations caused by the rapid variation of PV generation. Simulation results on an unbalance IEEE 123-bus system show that the proposed method can achieve similar performance as that of perfect physical-model-based approaches while being advantageous over other traditional methods.