Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “deep”

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 721 records · Page 40

Harness the power of atomistic modeling and deep learning in biofuel separation

Biofuels offer a remarkable, sustainable energy source for a future of clean energy. The development of efficient biofuel separation plays a crucial role in achieving cost-effective utilization of biofuel. In this chapter, we provide an overview of the recent advancements in atomistic-level modeling and deep learning in the rational design of novel, efficient biofuel separation. The fundamental principles of quantum and statistical mechanics are covered in appropriate detail to highlight their underlying differences in theory. The methodologies of several molecular representations and deep learning algorithms applicable to biofuel separation are briefly demonstrated as well. The applications, successes, and risks of employing density functional theory, ab initio molecular dynamics, classical molecular dynamics, and deep learning are provided to showcase their recent accomplishments in biofuel separation as well as potential improvements in both methodology and application. Lastly, a vision for the future growth of these methods is illustrated.

deep learning, artificial intelligence, biofuels, ↗

Building thermal load prediction through shallow machine learning and deep learning

Building thermal load prediction informs the optimization of cooling plant and thermal energy storage. Physics-based prediction models of building thermal load are constrained by the model and input complexity. In this study, we developed 12 data-driven models (7 shallow learning, 2 deep learning, and 3 heuristic methods) to predict building thermal load and compared shallow machine learning and deep learning. The 12 prediction models were compared with the measured cooling demand. It was found XGBoost (Extreme Gradient Boost) and LSTM (Long Short Term Memory) provided the most accurate load prediction in the shallow and deep learning category, and both outperformed the best baseline model, which uses the previous day's data for prediction. Then, we discussed how the prediction horizon and input uncertainty would influence the load prediction accuracy. Major conclusions are twofold: first, LSTM performs well in short-term prediction (1 h ahead) but not in long term prediction (24 h ahead), because the sequential information becomes less relevant and accordingly not so useful when the prediction horizon is long. Second, the presence of weather forecast uncertainty deteriorates XGBoost's accuracy and favors LSTM, because the sequential information makes the model more robust to input uncertainty. Training the model with the uncertain rather than accurate weather data could enhance the model's robustness. Finally, our findings have two implications for practice. First, LSTM is recommended for short-term load prediction given that weather forecast uncertainty is unavoidable. Second, XGBoost is recommended for long term prediction, and the model should be trained with the presence of input uncertainty.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Inverse design of hypoeutectoid pearlite steel microstructures using a deep learning and genetic algorithm optimization framework

Goal-oriented microstructure design in metallic materials is a challenging task due to complex structure-property relationships. Traditional experimental and computational approaches are time-intensive and economically inefficient, limiting their applicability for large-scale design space exploration. Here, in this work, we propose an end-to-end framework that integrates deep learning models with genetic optimization to design microstructures with targeted mechanical properties. Deep learning models enable accurate forward design, while their integration with genetic optimization enables efficient inverse design within a few hours, compared to days or weeks using conventional finite element simulations. The framework combines experimental characterization and finite element modeling to analyze the influence of microstructural features on the mechanical behavior of hypoeutectoid steels. Data from both experiments and simulations are used to train the deep learning models. To demonstrate its effectiveness, we apply the framework to 0.63% carbon steel with proeutectoid ferrite and pearlite phases, commonly used in industrial applications. In this study, 2D microstructures were used for modeling, selected primarily for computational efficiency and to establish proof of concept. The framework successfully optimizes microstructures for targeted yield strength, ultimate strength, and stress concentration factors while significantly reducing computational time. Beyond hypoeutectoid steels, this scalable framework can be extended to other material systems and integrated with additive manufacturing, offering an efficient approach for accelerating microstructure design for specific engineering applications.

ConvLSTM↗

Deep reinforcement learning control of hydraulic fracturing

Hydraulic fracturing is a technique to extract oil and gas from shale formations, and obtaining a uniform proppant concentration along the fracture is key to its productivity. Recently, various model predictive control schemes have been proposed to achieve this objective. But such controllers require an accurate and computationally efficient model which is difficult to obtain given the complexity of the process and uncertainties in the rock formation properties. In this article, we design a model-free data-based reinforcement learning controller which learns an optimal control policy through interactions with the process. Deep reinforcement learning (DRL) controller is based on the Deep Deterministic Policy Gradient algorithm that combines Deep-Q-network with actor-critic framework. In addition, we utilize dimensionality reduction and transfer learning to quicken the learning process. We show that the controller learns an optimal policy to obtain uniform proppant concentration despite the complex nature of the process while satisfying various input constraints.

42 ENGINEERING↗

Comprehensive assessment of deep reinforcement learning approaches for economic dispatch in nuclear-driven microgrids

As the electrical grid integrates more variable renewable energy sources such as wind and solar, the demand for distributed and flexible systems to address this increased variability becomes critical. Nuclear-driven microgrids provide a promising solution by offering stable generation to complement intermittent renewables, ensuring grid reliability and operating efficiency. This paper proposes a recurrent deep reinforcement learning framework for optimal economic dispatch in a nuclear-powered microgrid integrating renewable energy sources, small modular reactors, battery storage systems, and balance-of-plant dynamics. A three-agent control architecture is developed, where demand and renewable energy agents act as forecasters, and a reinforcement learning-based dispatch agent performs real-time energy allocation. A nonlinear programming formulation is first used to generate an optimal baseline for benchmarking. The proposed dispatch controller, based on Proximal Policy Optimization enhanced with Long Short-Term Memory networks, exploits temporal correlations in system dynamics by taking advantage of the time series used as inputs to improve policy robustness under uncertainty. Comparative analysis against established deep reinforcement learning methods, including Proximal Policy Optimization with a feedforward architecture, Soft Actor-Critic, and Twin Delayed Deep Deterministic Policy Gradient, demonstrates superior performance. Numerical results indicate that the proposed controller achieves a 0.39% cost reduction relative to the nonlinear programming benchmark and outperforms other learning-based methods by generating additional revenue of up to 0.35%. All reinforcement learning controllers compute dispatch actions in less than 0.3 s, resulting in a computational speedup of more than three orders of magnitude over the nonlinear programming baseline. The findings of this paper highlight their applicability for real-time operation and control in nuclear-integrated microgrids under volatile operating conditions.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Deep learning–based digital twins for heat pumps

Heat pumps are effective cooling and heating appliances to save energy in buildings. However, traditional heat pump models are challenging to integrate with building demands in a co-simulation environment because of the nonlinear thermodynamics of refrigerants. Developing digital twin representatives for heat pumps capable of faster calculations with good accuracy is desirable. This study aimed to establish a generic deep learning–based digital twin for heat pumps with a large amount of high-fidelity data. Two refrigerants for two different heat pumps were considered: an air source heat pump with refrigerant R-410A, an air source heat pump with refrigerant CO 2 , a water source heat pump with refrigerant R-410A, and a water source heat pump with refrigerant CO 2 . Furthermore, results showed that the deep learning (long short-term memory) models effectively represented these four heat pumps as a digital twin: (a) accuracy for training and testing showed smaller than 0.02 for heating electricity and heating demands, and (b) the digital twins showed good consistency with original data for heating electricity and heating demands (root mean square errors of less than 0.12 W and 0.19 W, respectively). Therefore, deep learning–based heat pump models can be used in the co-simulation of building mechanical systems.

Air source heat pump↗

Development of lean, efficient, and fast physics-framed deep-learning-based proxy models for subsurface carbon storage

In this work, we present deep-learning-based surrogate models for CCUS developed with four different algorithms and a physics-framed two-phase flow problem involving displacement of water by CO 2 . The deep-learning models were trained using 3D datasets describing the pressure plume, CO 2 saturation plume, and water extraction rate generated by numerical simulation. The hyperparameters defining the architecture of the neural networks were optimized to determine the slimmest network size and training parameters that give the most efficient performance at the least training cost. To develop a robust model that closely mimics the governing physical laws, the discretized form of the two-phase fluid transport equation was used to formulate the supervised deep-learning task. The algorithms investigated in this study predicted the data to above 95% accuracy, with the multi-layer perceptron model demonstrating the best performance by balancing training speed, prediction time, and prediction accuracy with lean network capacity. Furthermore, the surrogate models simultaneously predict reservoir pressure and CO 2 saturation in every grid block, including the surface well extraction rate and bottomhole pressure, at all simulation times for a given static model realization in just a few seconds on a standard desktop computer. A key outcome of this study is that limits can be placed on network design parameters to avoid over designing neural networks, with associated efficiencies in training and prediction times. This is very useful because large volumes of data may be generated in CCUS projects and over-design of neural network architectures imposes penalties that are antithetical to the goal of near-real time forecasting.

58 GEOSCIENCES↗

Unveiling mesoscopic structures in distorted lamellar phases through deep learning-based small angle neutron scattering analysis

Hypothesis: The formation of distorted lamellar phases, distinguished by their arrangement of crumpled, stacked layers, is frequently accompanied by the disruption of long-range order, leading to the formation of interconnected network structures commonly observed in the sponge phase. Nevertheless, traditional scattering functions grounded in deterministic modeling fall short of fully representing these intricate structural characteristics. Our hypothesis posits that a deep learning method, in conjunction with the generalized leveled wave approach used for describing structural features of distorted lamellar phases, can quantitatively unveil the inherent spatial correlations within these phases. Experiments and Simulations: This report outlines a novel strategy that integrates convolutional neural networks and variational autoencoders, supported by stochastically generated density fluctuations, into a regression analysis framework for extracting structural features of distorted lamellar phases from small angle neutron scattering data. To evaluate the efficacy of our proposed approach, we conducted computational accuracy assessments and applied it to the analysis of experimentally measured small angle neutron scattering spectra of AOT surfactant solutions, a frequently studied lamellar system. Findings: The findings unambiguously demonstrate that deep learning provides a dependable and quantitative approach for investigating the morphology of wide variations of distorted lamellar phases. It is adaptable for deciphering structures from the lamellar to sponge phase including intermediate structures exhibiting fused topological features. In conclusion, this research highlights the effectiveness of deep learning methods in tackling complex issues in the field of soft matter structural analysis and beyond.

36 MATERIALS SCIENCE↗

Effects of carbon dioxide sequestration on California margin deep-sea foraminiferal assemblages

Abstract Deep-sea sequestration of CO2 is being considered as a possible mitigation tool to decrease atmospheric CO2 concentrations and its associated negative effects. This study investigated potential effects of liquid carbon dioxide (CO2) injection on deep-sea foraminiferal assemblages. Foraminifera are ideal for this ecological impact investigation because of differing test composition (calcareous and non-calcareous) and thickness, and diverse epifaunal and infaunal depth preferences. The experiment was conducted on August-September 2003, at 3600 m off the coast of Monterey Bay, California, aboard the R/V Western Flyer using the ROV Tiburon. The pH of the site was monitored throughout the experiment. Sediment push-cores were collected (both from the experimental and control sites) and stained to distinguish live (stained) from dead (unstained) individuals. Effects of CO2 injection on assemblages have been tracked both vertically (to 10 cm depth below sea floor) and horizontally (up to 10 m from CO2 injection sites), as well as between live and dead individuals. Within corrals (containing the injected CO2) and their underlying sediments, severe pH changes (near 4.0 units) were recorded. This compares with a record of small average reductions in ocean pH (-0.05 units) combined with large episodic excursions (-1.7 units) over the experimental area due to the injection of CO2. Exposure to this gradient of low pH caused increased mortality and dissolution of calcareous forms within corrals, as far as 5 m from the injection site, and to at least 10 cm depth in the sediments. This experiment revealed several major effects of CO2 injection on foraminiferal assemblages in surficial sediments: 1) total number of foraminifera in a sample decreases; 2) foraminiferal species richness decreases in both stained and unstained specimens; and 3) relative percentage of stained (live) forms in the remaining tests increases. Down-core trends (to 10 cm below sea floor) have revealed: 1) percent agglutinated forms decline and calcareous forms increase with depth; 2) agglutinated species richness decreases with depth; and 3) experimental core assemblages become increasingly similar with depth to those in control cores not subjected to CO2 injection. These results imply almost complete initial mortality and dissolution in the upper 10 cm throughout the corrals following liquid CO2 injection. Since calcareous foraminifera represent more than 50% of the total assemblages, this clearly indicates that emplacement of CO2 will result in negative effects to diversity and survivorship of the deep-sea benthic meiofauna.

54 ENVIRONMENTAL SCIENCES↗

Coupled deep-mantle carbon-water cycle: Evidence from lower-mantle diamonds

Diamonds form in a variety of environments between subducted crust, lithospheric and deep mantle. Recently, deep source diamonds with inclusions of the high-pressure H 2 O-phase ice-VII were discovered. By correlating the pressures of ice-VII inclusions with those of other high-pressure inclusions, we assess quantitatively the pressures and temperatures of their entrapment. We show that the ice-VII-bearing diamonds formed at depths down to 800 ± 60 km but at temperatures 200–500 K below average mantle temperature that match the pressure-temperature conditions of decomposing dense hydrous mantle silicates. Furthermore, our work presents strong evidence for coupled recycling of water and carbon in the deep mantle based on natural samples.

58 GEOSCIENCES↗

Reproducibility of Radiokrypton in Deep Desert Aquifers: Insights from a Decade of Research

Great technical advances have been achieved since the first atom-trap trace analysis (ATTA) -based radiokrypton application in Egypt, where 1 Myr old groundwater was discovered. Beyond advances in ATTA measurement capabilities, including reduction in sample size, analysis duration, and analytical uncertainty, major progress has been achieved over the past two decades in the sample collection and preparation techniques. These advances paved the expansion of ATTA-based noble gas applications to many other aquifers worldwide, illuminating the nature and flow pattern of deep groundwater systems. While the potential of this new analytical technique for old groundwater dating is well recognized, another important aspect yet to be examined is the reproducibility of radiokrypton in aquifers over time, i.e., how representative is a discrete groundwater sample, collected at a specific time and location, for the natural groundwater system? The likelihood of a negative answer is increased by flow-field disturbance in aquifers following massive groundwater abstraction. Here, in this work, we present repeated 81 Kr sampling and measurements in twenty-one sites over Israel, mostly of deep (up to 1 km) wells tapping confined aquifers in the arid to hyperarid Negev desert. The results demonstrate that radiokrypton measurements are indeed reproducible, even in cases where samples were collected as long as nine years apart and from highly productive (∼1 Mm 3 /yr order) pumping wells. Furthermore, many of the repeated measurements in this study (17 out of the 21 sites) were conducted with different ATTA Instruments in two different laboratories using slightly different sampling, preparation, and analysis techniques, yet with an overall good agreement. The consistency in the ATTA-based 81 Kr-dating results over time highlights the robustness of this state-of-the-art technique as a tool to unravel groundwater flow patterns and encourages further applications to many other yet-to-be-explored deep aquifers.

atom-trap trace analysis↗

Stability of a Mixed–Valence Hydrous Iron–Rich Oxide: Implications for Water Storage and Dynamics in the Deep Lower Mantle

Incorporation of water into mantle compositions can have significant effects on the phase relations in the systems. Here, we synthesized an iron-rich hexagonal hydrous phase (referred to as "HH1-phase") under the high pressure-temperature (P-T) conditions of the deep lower mantle and determined the crystal structure of the HH1-phase at 79 GPa using the multigrain crystallography method. The chemical formula obtained was Fe 12.76 O 18 H x (x ~ to 4.5) in the Fe-O-H system. To demonstrate the role of HH1-phase for water storage in multicomponent systems relevant to mantle compositions, we investigated the stability of HH1-phase in both MgO-rich pyrolitic and SiO 2 -rich basaltic compositions. Our results indicate that the HH1-phase serves as major water storage in a pyrolitic composition, whereas the Al-rich CaCl 2 -type δ-phase and SiO 2 phase are major water storage phases in a SiO 2 -rich basaltic composition. Incorporation of considerable amounts of SiO 2 , MgO, and Al 2 O 3 into the HH1-phase expands its stability field from 98 GPa in the Fe-Al-O-H system to at least 108 GPa (corresponding to similar to 2,400 km depth) in the Mg-Si-Al-Fe-O-H system. Plumes of hot upwelling rock rooted at the base of the lower mantle have been proposed as a possible origin of hotspot volcanoes. The hydrous Fe-rich HH1-phase, if included into the material of upwelling plumes, will decompose on its rising to the upper part of the lower mantle and release water. Our results should provide constraints on water storage in the deep lower mantle and have implications for deep mantle dynamics.

58 GEOSCIENCES↗

Aragonite saturation horizon variability along North Pacific seamounts and implications for deep-sea coral reefs

The 2014 discovery of living deep-sea coral reefs along the Northwest Hawaiian Islands (NWHI) and lower Emperor Seamount Chain (ESC), despite the North Pacific’s shallow aragonite saturation horizon (ASH) and high CaCO3 dissolution rates, underscores the need to understand the local seawater chemistry that allows these reefs to persist. We investigated seawater carbonate chemistry using discrete samples along the NWHI and ESC from two cruises ~1 year apart (08/26/21 – 09/26/21, 09/09/22 – 10/24/22). Across the two cruises, ASH depth difference ranged from 15 to 77 m. Since the Pacific ASH shoals by 1–2 m yr?¹, this long-term trend cannot explain the magnitude of ASH change observed between cruises. Temperature-salinity plots similarly indicate no major shifts in intermediate water masses that could account for these changes. Instead, ASH depth variability was primarily governed by localized biogeochemical processes, namely changes in intermediate water respiration and CaCO3 dissolution. Indicators for dissolution (TA*) and respiration (AOU) suggest changes in ASH depth were driven by changes in dissolution at the northern- and southern-most sites, whereas respiration exerted stronger control at central sites. Combining 2021 and 2022 data with data from 2014 – 2019 (excluding 2018) revealed high interannual ASH variability, by as much as >200 m at one site. Deep-sea coral reefs across the NWHI and ESC currently reside close to the ASH depth and likely experience seasonal and interannual shifts between under- and supersaturation. As ocean acidification progresses however, persistent exposure to undersaturated seawater could further threaten these critical deep-sea ecosystems.

coral reefs↗

Isochronal annealing study of low energy electron irradiated Al-doped p-type 6H silicon carbide with deep level transient spectroscopy

Al doped p-type 6H silicon carbide was irradiated by low energy electrons to create primary defects. Two deep levels at E{sub V}+0.36 eV and E{sub V}+0.81 eV were created by this irradiation. Isochronal annealing study was carried out on the electron irradiated sample to investigate the annealing out of the two primary defects and the creation of thermal annealing-induced secondary defects. Four more deep hole traps (0.45, 0.56, 0.74, and 0.71 eV above the valence band) were formed during the whole annealing process up to a temperature of 1600 deg. C. All the electron irradiated deep level defects were annealed out after the 1600 deg. C annealing.

36 MATERIALS SCIENCE↗

Deep traps in nonpolar m-plane GaN grown by ammonia-based molecular beam epitaxy

Deep level defects in nonpolar m-plane GaN grown by ammonia-based molecular beam epitaxy were characterized using deep level transient spectroscopy (DLTS) and deep level optical spectroscopy (DLOS) and compared with polar c-plane GaN that was grown simultaneously in the same growth run. Significant differences in both the levels present and their concentrations were observed upon comparison of both growth orientations. DLTS revealed electron traps with activation energies of 0.14 eV, 0.20 eV, and 0.66 eV in the m-plane material, with concentrations that were {approx}10-50 x higher than traps of similar activation energies in the c-plane material. Likewise, DLOS measurements showed {approx}20 x higher concentrations of both a C{sub N} acceptor-like state at E{sub C} - 3.26 eV, which correlates with a high background carbon concentration observed by secondary ion mass spectroscopy for the m-plane material [A. Armstrong, A. R. Arehart, B. Moran, S. P. DenBaars, U. K. Mishra, J. S. Speck, and S. A. Ringel, Appl. Phys. Lett. 84, 374 (2004)], and the V{sub Ga}-related state level at E{sub C} - 2.49 eV, which is consistent with an enhanced yellow luminescence observed by photoluminescence. The findings suggest a strong impact of growth dynamics on the incorporation of impurities and electrically active native point defects as a function of GaN growth plane polarity.

36 MATERIALS SCIENCE↗

Investigation on origin of Z{sub 1/2} center in SiC by deep level transient spectroscopy and electron paramagnetic resonance

The Z{sub 1/2} center in n-type 4H-SiC epilayers-a dominant deep level limiting the carrier lifetime-has been investigated. Using capacitance versus voltage (C-V) measurements and deep level transient spectroscopy (DLTS), we show that the Z{sub 1/2} center is responsible for the carrier compensation in n-type 4H-SiC epilayers irradiated by low-energy (250 keV) electrons. The concentration of the Z{sub 1/2} defect obtained by C-V and DLTS correlates well with that of the carbon vacancy (V{sub C}) determined by electron paramagnetic resonance, suggesting that the Z{sub 1/2} deep level originates from V{sub C}.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Origin and annealing of deep-level defects in GaNAs grown by metalorganic vapor phase epitaxy

Deep-level defects were investigated by deep level transient spectroscopy on the as-grown and annealed GaNAs layers of various nitrogen (N) contents. The unintentionally doped (uid) GaNAs layers were grown by metalorganic vapor phase epitaxy with N = 1.4%, 2.0%, 2.2%, and 2.4% on GaAs substrate. The possible origin and evolution of the deep-level defects upon annealing were analyzed with the use of the GaNAs band gap diagram concept [Kudrawiec et al., Appl. Phys. Lett. 101, 082109 (2012)], which assumes that the activation energy of donor traps decreases with N-related downward shift of the conduction band. On the basis of this diagram and in comparison with previous results, the N-related traps were associated with (N−As){sub As} or (N−N){sub As} split interstitials. It was also proposed that one of the electron traps and the hole trap, lying at the same level position in the bandgap of the annealed uid-GaNAs layers, can both act as one generation-recombination center partially responsible for poor optical properties of this alloy.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Donor and double-donor transitions of the carbon vacancy related EH{sub 6∕7} deep level in 4H-SiC

Using medium- and high-resolution multi-spectra fitting of deep level transient spectroscopy (DLTS), minority carrier transient spectroscopy (MCTS), optical O-DLTS and optical-electrical (OE)-MCTS measurements, we show that the EH{sub 6∕7} deep level in 4H-SiC is composed of two strongly overlapping, two electron emission processes with thermal activation energies of 1.49 eV and 1.58 eV for EH{sub 6} and 1.48 eV and 1.66 eV for EH{sub 7}. The electron emission peaks of EH{sub 7} completely overlap while the emission peaks of EH{sub 6} occur offset at slightly different temperatures in the spectra. OE-MCTS measurements of the hole capture cross section σ{sub p0}(T) in p-type samples reveal a trap-Auger process, whereby hole capture into the defect occupied by two electrons leads to a recombination event and the ejection of the second electron into the conduction band. Values of the hole and electron capture cross sections σ{sub n}(T) and σ{sub p}(T) differ strongly due to the donor like nature of the deep levels and while all σ{sub n}(T) have a negative temperature dependence, the σ{sub p}(T) appear to be temperature independent. Average values at the DLTS measurement temperature (∼600 K) are σ{sub n2+}(T) ≈ 1 × 10{sup −14} cm{sup 2}, σ{sub n+}(T) ≈ 1 × 10{sup −14} cm{sup 2}, and σ{sub p0}(T) ≈ 9 × 10{sup −18} cm{sup 2} for EH{sub 6} and σ{sub n2+}(T) ≈ 2 × 10{sup −14} cm{sup 2}, σ{sub n+}(T) ≈ 2 × 10{sup −14} cm{sup 2}, σ{sub p0}(T) ≈ 1 × 10{sup −20} cm{sup 2} for EH{sub 7}. Since EH{sub 7} has already been identified as a donor transition of the carbon vacancy, we propose that the EH{sub 6∕7} center in total represents the overlapping first and second donor transitions of the carbon vacancy defects on both inequivalent lattice sites.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗