Search NASA⌕ Search

SEARCH · Search NASA

Results for “Subsurface Simulations”

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 559 records · Page 31

Investigation of CO2 plume behavior for a large-scale pilot test of geologic carbon storage in a saline formation

The hydrodynamic behavior of carbon dioxide (CO{sub 2}) injected into a deep saline formation is investigated, focusing on trapping mechanisms that lead to CO{sub 2} plume stabilization. A numerical model of the subsurface at a proposed power plant with CO{sub 2} capture is developed to simulate a planned pilot test, in which 1,000,000 metric tons of CO{sub 2} is injected over a four-year period, and the subsequent evolution of the CO{sub 2} plume for hundreds of years. Key measures are plume migration distance and the time evolution of the partitioning of CO{sub 2} between dissolved, immobile free-phase, and mobile free-phase forms. Model results indicate that the injected CO{sub 2} plume is effectively immobilized at 25 years. At that time, 38% of the CO{sub 2} is in dissolved form, 59% is immobile free phase, and 3% is mobile free phase. The plume footprint is roughly elliptical, and extends much farther up-dip of the injection well than down-dip. The pressure increase extends far beyond the plume footprint, but the pressure response decreases rapidly with distance from the injection well, and decays rapidly in time once injection ceases. Sensitivity studies that were carried out to investigate the effect of poorly constrained model parameters permeability, permeability anisotropy, and residual CO{sub 2} saturation indicate that small changes in properties can have a large impact on plume evolution, causing significant trade-offs between different trapping mechanisms.

54 ENVIRONMENTAL SCIENCES↗

Permafrost thermal conditions are sensitive to shifts in snow timing

Abstract Changes in snow precipitation at high latitudes can significantly affect permafrost thermal conditions and thaw depth, potentially exposing more carbon-laden soil to microbial decomposition. A fully coupled process-based surface/subsurface thermal hydrology model with surface energy balance is used to analyze the impact of intra-annual variability in snow on permafrost thermal regime and the active layer thickness. In the four numerical scenarios considered, simulations were forced by the same meteorological data, except the snow precipitation, which was systematically altered to change timing of snowfall. The scenarios represent subtle shifts in snow timing, but the snow onset/melt days, the end of winter snowpack depth, and total annual snow precipitation are unchanged among scenarios. The simulations show that small shifts in the timing of snow accumulation can have significant effects on subsurface thermal conditions leading to active layer deepening and even talik formation when snowfall arrives earlier in the winter. The shifts in snow timing have a stronger impact on wetter regions, especially soil underneath small ponds, as compared to drained regions. This study highlights the importance of understanding potential changes in winter precipitation patterns for reliable projections of active-layer thickness in a changing Arctic climate.

54 ENVIRONMENTAL SCIENCES↗

Physical and Numerical Models of Pore-Scale Trapping of CO2

Numerical simulations of deep geological carbon sequestration are being used to design field-scale implementations, predict their long-term performance, and evaluate associated risks. However, conventional field-scale simulators are unable to explicitly resolve the pore-scale processes that exert significant control on the migration and fate of supercritical CO2 in the subsurface. A combination of physical and numerical experiments with explicit pore-scale resolution are being performed in order to identify conditions under which continuum model approximations may fail, and to propose alternative simulation approaches that can be used in such cases.

54 ENVIRONMENTAL SCIENCES↗

Generation of Subsurface Voids, Incubation Effect, and Formation of Nanoparticles in Short Pulse Laser Interactions with Bulk Metal Targets in Liquid: Molecular Dynamics Study

The ability of short pulse laser ablation in liquids to produce clean colloidal nanoparticles and unusual surface morphology has been employed in a broad range of practical applications. In this paper, we report the results of large-scale molecular dynamics simulations aimed at revealing the key processes that control the surface morphology and nanoparticle size distributions by pulsed laser ablation in liquids. The simulations of bulk Ag targets irradiated in water are performed with an advanced computational model combining a coarse-grained representation of liquid environment and an atomistic description of laser interaction with metal targets. For the irradiation conditions that correspond to the spallation regime in vacuum, the simulations predict that the water environment can prevent the complete separation of the spalled layer from the target, leading to the formation of large subsurface voids stabilized by rapid cooling and solidification. The subsequent irradiation of the laser-modified surface is found to result in a more efficient ablation and nanoparticle generation, thus suggesting the possibility of the incubation effect in multipulse laser ablation in liquids. The simulations performed at higher laser fluences that correspond to the phase explosion regime in vacuum reveal the accumulation of the ablation plume at the interface with the water environment and the formation of a hot metal layer. The water in contact with the metal layer is brought to the supercritical state and provides an environment suitable for nucleation and growth of small metal nanoparticles from metal atoms emitted from the hot metal layer. The metal layer itself has limited stability and can readily disintegrate into large (tens of nanometers) nanoparticles. The layer disintegration is facilitated by the Rayleigh–Taylor instability of the interface between the higher density metal layer decelerated by the pressure from the lighter supercritical water. The nanoparticles emerging from the layer disintegration are rapidly cooled and solidified due to the interaction with water environment, with a cooling rate of ~2 × 10 12 K/s observed in the simulations. The computational prediction of two distinct mechanisms of nanoparticle formation yielding nanoparticles with different characteristic sizes provides a plausible explanation for the experimental observations of bimodal nanoparticle size distributions in laser ablation in liquids. The ultrahigh cooling and solidification rates suggest the possibility for generation of nanoparticles featuring metastable phases and highly nonequilibrium structures.

Chemistry↗

Model Development for Thermal-Hydrology Simulations of a Full-Scale Heater Experiment in Opalinus Clay

Disposal of commercial spent nuclear fuel in a geologic repository is studied. In situ heater experiments in underground research laboratories provide a realistic representation of subsurface behavior under disposal conditions. Here, this study describes process model development and modeling analysis for a full-scale heater experiment in opalinus clay host rock. The results of thermal-hydrology simulation, solving coupled nonisothermal multiphase flow, and comparison with experimental data are presented. The modeling results closely match the experimental data.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

3D Realistic Modeling of the Interaction of Quiet-Sun Magnetic Fields with the Chromosphere

High-resolution observations and 3D simulations suggest that a local dynamo operates near the surface and produces ubiquitous small-scale magnetic elements, thus contributing to the magnetic carpet in the photosphere and to the magnetic structure and dynamics of the solar atmosphere. It appears that the traditional mechanisms of chromospheric energy and mass transport by acoustic waves and shocks are likely to play a secondary role; instead, the primary drivers in the energetics and dynamics of the chromosphere and transition region are small-scale, previously unresolved, quiet-Sun magnetic fields. These fields appear as ubiquitous, rapidly changing (on the scale of a few seconds), tiny magnetic loops and magnetized vortex tubes. Questions then arise about their origin and dynamics in the chromosphere, their links to magnetic fields in the photosphere, and their role in the energy storage and exchange between subsurface layers and the chromosphere. In the talk we will present results of 3D radiative MHD simulations obtained with the StellarBox code and discuss the energetics and dynamical interlinks between the subphotospheric layers and low chromosphere, their effects on the structure of the chromosphere, and signatures of the fine-scale magnetic features in high-resolution spectro-polarimetric observations.

Quie-Sun↗

Lattice Boltzmann simulation of immiscible fluid displacement in porous media: Homogeneous versus heterogeneous pore network

Injection of anthropogenic carbon dioxide (CO{sub 2}) into geological formations is a promising approach to reduce greenhouse gas emissions into the atmosphere. Predicting the amount of CO{sub 2} that can be captured and its long-term storage stability in subsurface requires a fundamental understanding of multiphase displacement phenomena at the pore scale. In this paper, the lattice Boltzmann method is employed to simulate the immiscible displacement of a wetting fluid by a non-wetting one in two microfluidic flow cells, one with a homogeneous pore network and the other with a randomly heterogeneous pore network. We have identified three different displacement patterns, namely, stable displacement, capillary fingering, and viscous fingering, all of which are strongly dependent upon the capillary number (Ca), viscosity ratio (M), and the media heterogeneity. The non-wetting fluid saturation (S{sub nw}) is found to increase nearly linearly with logCa for each constant M. Increasing M (viscosity ratio of non-wetting fluid to wetting fluid) or decreasing the media heterogeneity can enhance the stability of the displacement process, resulting in an increase in S{sub nw}. In either pore networks, the specific interfacial length is linearly proportional to S{sub nw} during drainage with equal proportionality constant for all cases excluding those revealing considerable viscous fingering. Our numerical results confirm the previous experimental finding that the steady state specific interfacial length exhibits a linear dependence on S{sub nw} for either favorable (M ≥ 1) or unfavorable (M < 1) displacement, and the slope is slightly higher for the unfavorable displacement.

54 ENVIRONMENTAL SCIENCES↗

Adatom-Driven Oxygen Intermixing during the Deposition of Oxide Thin Films by Molecular Beam Epitaxy

Thin film deposition from the vapor phase is a complex process involving adatom adsorption, movement, and incorporation into the growing film. While simulations have been able to capture key aspects of these processes to reproduce film morphology, these models typically ignore any subsurface diffusion that occurs. The results of these diffusion and intermixing processes are often observed experimentally, but cannot be explained by invoking bulk lattice diffusion at the deposition temperature. In this work, we present quantitative experimental data that reveals anion and cation intermixing over long length scales during the deposition of epitaxial Fe 2 O 3 and Cr 2 O 3 films and heterostructures by oxygen-plasma-assisted molecular beam epitaxy. We track this diffusion by incorporating well-defined tracer layers containing 18 O and/or 57 Fe, and measure their redistribution on the nanometer scale with atom probe tomography. Molecular dynamics simulations suggest potential intermixing events, which are then examined via nudged elastic band calculations. These reveal that adatoms on the film surface act to “pull up” subsurface O and Fe. Subsequent ring-like rotation mechanisms involving both adatom and subsurface anions then facilitate their mixing. A simple model incorporating these events qualitatively agrees with the experimental intermixing results. In addition to film deposition, these intermixing mechanisms may to be operant during other surface-mediated processes such as heterogeneous catalysis and corrosion.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Predicting Flow in Fracture Networks With Quantum Algorithms

Uncertainty quantification plays a crucial role in the modeling of subsurface flow. For instance, uncertainties in the properties of geologic fracture networks significantly impact flow, requiring numerous simulations to accurately estimate quantities of interest. However, each simulation is computationally expensive because it requires solving a large linear system to capture features that involve both small and large fractures. An example is in percolation, where the interaction of many small fractures (which cumulatively can have a large surface area) with the rock matrix must be modeled precisely. Quantum computing is an emerging tool with the potential to address this issue. Quantum algorithms offer a significant speedup in solving linear systems, achieving efficiencies that are challenging to match with classical approaches. These classical approaches include direct solvers, such as LU decomposition, and iterative methods, notably preconditioned conjugate gradient, commonly used in subsurface modeling to solve large sparse systems. However, applying quantum algorithms to geologic fracture flow requires careful attention to algorithmic and problem-specific constraints to fully realize this quantum advantage. In this work we describe a quantum algorithm for generalized Monte Carlo applications with a quadratic speedup over the classical approaches which can be combined with the quantum speedup, currently under investigation, for solving quantum linear systems for subsurface flow. We show that for quantum algorithms the computational cost of estimating a quantity of interest for a statistical ensemble of networks is roughly the same as that of a single realization, essentially implying that one can get uncertainty quantification for free.

58 GEOSCIENCES↗

Integrating the Interconnections Between Groundwater and Land Surface Processes Through the Coupled NASA Land Information System and ParFlow Environment

Understanding the interactions between the atmosphere, the land, and the subsurface is fundamental to hydrology and is critical for a better assessment of the impacts of climate change and human management on hydrological systems. However, many land surface models simplify the subsurface hydrology and thereby these interactions. In this study, we couple the land surface model Noah-MP included in the NASA Land Information System (LIS) with the integrated hydrologic model ParFlow (ParFlow-LIS) using the Earth System Modeling Framework (ESMF) and the National United Operational Prediction Capability (NUOPC). This coupling improves the simulation of water and energy cycle processes by adding the three-dimensional variably saturated and heterogeneous flow in the subsurface using sophisticated and nonlinear physics-based equations as well as the advances in satellite remote sensing-based data assimilation of the land surface, thereby benefiting the integrated hydrologic modeling and data assimilation community. We use the High Plains aquifer, located in the central United States, as a testbed to evaluate the coupled ParFlow-LIS system. The new ParFlow-LIS system accounts for the effects of topographically driven flows on the land surface, producing more fine-scale patterns of land surface states and fluxes than standalone LIS. In addition, ParFlow-LIS enables the consideration of the effect of subsurface water storage on evapotranspiration. This is particularly important in areas and times with dry soils, such as during drought conditions or in the presence of a cone of depression due to pumping.

land surface modeling↗

Understanding Discrete Fracture Networks Through Spectral Graph Theory

Discrete Fracture Network models (DFNs) are used to simulate fluid flow and particle transport through fracture networks in low permeability rock. Understanding these processes are essential in many subsurface applications, such as environmental restoration of contaminated fractured media, CO 2 sequestration, detection of low-level nuclear tests, and hydrocarbon extraction. Compared with other models, DFNs allow for incorporation of a wider range of network characteristics but have substantially greater computation cost. These networks can be represented with graphs, allowing the use of graph theory tools to study the networks. I used Python to simulate flow and transport on a range of DFNs and analyzed these networks using methods from network analysis and spectral graph theory. My purpose was to find ways to gain insight about flow and transport on DFNs using these graph representations, bypassing the computationally intensive meshing typically required. My work is still in progress, but I have discovered several interesting trends and patterns that I believe could be useful towards my goal. If I am able to bring these results to fruition, they will aid subsurface geologists in extracting flow and transport information about fracture networks more efficiently.

54 ENVIRONMENTAL SCIENCES↗

Climate-Soil-Vegetation Control on Groundwater Table Dynamics and its Feedbacks in a Climate Model

Among the three dynamically linked branches of the water cycle, including atmospheric, surface, and subsurface water, groundwater is the largest reservoir and an active component of the hydrologic system. Because of the inherent slow response time, groundwater may be particularly relevant for long time-scale processes such as multi-years or decadal droughts. This study uses regional climate simulations with and without surface water – groundwater interactions for the conterminous U.S. to assess the influence of climate, soil, and vegetation on groundwater table dynamics, and its potential feedbacks to regional climate. Analysis shows that precipitation has a dominant influence on the spatial and temporal variations of groundwater table depth (GWT). The simulated GWT is found to decrease sharply with increasing precipitation. Our simulation also shows some distinct spatial variations that are related to soil porosity and hydraulic conductivity. Vegetation properties such as minimum stomatal resistance, and root depth and fraction are also found to play an important role in controlling the groundwater table. Comparing two simulations with and without groundwater table dynamics, we find that groundwater table dynamics mainly influences the partitioning of soil water between the surface (0 – 0.5 m) and subsurface (0.5 – 5 m) rather than total soil moisture. In most areas, groundwater table dynamics increases surface soil moisture at the expense of the subsurface, except in regions with very shallow groundwater table. The change in soil water partitioning between the surface and subsurface is found to strongly correlate with the partitioning of surface sensible and latent heat fluxes. The evaporative fraction (EF) is generally higher during summer when groundwater table dynamics is included. This is accompanied by increased cloudiness, reduced diurnal temperature range, cooler surface temperature, and increased cloud top height. Although both convective and non-convective precipitation are enhanced, the higher EF changes the partitioning to favor more non-convective precipitation, but this result could be sensitive to the convective parameterization used. Compared to simulations without groundwater table dynamics, the dry bias in the summer precipitation is slightly reduced over the central and eastern U.S. Groundwater table dynamics can provide important feedbacks to atmospheric processes, and these feedbacks are stronger in regions with deeper groundwater table, because the interactions between surface and subsurface are weak when the groundwater table is deep. This increases the sensitivity of surface soil moisture to precipitation anomalies, and therefore enhances land surface feedbacks to the atmosphere through changes in soil moisture and evaporative fraction. By altering the groundwater table depth, land use change and groundwater withdrawal can alter land surface response and feedback to the climate system.

54 ENVIRONMENTAL SCIENCES↗

Functional approach to exploring climatic and landscape controls of runoff generation: 1. Behavioral constraints on runoff volume

Abstract Inspired by the Dunne diagram, the climatic and landscape controls on the partitioning of annual runoff into its various components (Hortonian and Dunne overland flow and subsurface stormflow) are assessed quantitatively, from a purely theoretical perspective. A simple distributed hydrologic model has been built sufficient to simulate the effects of different combinations of climate, soil, and topography on the runoff generation processes. The model is driven by a sequence of simple hypothetical precipitation events, for a large combination of climate and landscape properties, and hydrologic responses at the catchment scale are obtained through aggregation of grid‐scale responses. It is found, first, that the water balance responses, including relative contributions of different runoff generation mechanisms, could be related to a small set of dimensionless similarity parameters. These capture the competition between the wetting, drying, storage, and drainage functions underlying the catchment responses, and in this way, provide a quantitative approximation of the conceptual Dunne diagram. Second, only a subset of all hypothetical catchment/climate combinations is found to be “behavioral,” in terms of falling sufficiently close to the Budyko curve, describing mean annual runoff as a function of climate aridity. Furthermore, these behavioral combinations are mostly consistent with the qualitative picture presented in the Dunne diagram, indicating clearly the commonality between the Budyko curve and the Dunne diagram. These analyses also suggest clear interrelationships amongst the “behavioral” climate, soil, and topography parameter combinations, implying these catchment properties may be constrained to be codependent in order to satisfy the Budyko curve.

Li, Hong‐Yi↗

Reaction and diffusion at the reservoir/shale interface during CO 2 storage: Impact of geochemical kinetics

Here we use a reactive diffusion model to investigate what happens to CO 2 injected into a subsurface sandstone reservoir capped by a chlorite- and illite-containing shale seal. The calculations simulate reaction and transport of supercritical (SC) CO 2 at 348.15 K and 30 MPa up to 20,000 a. Given the low shale porosity (5%), chemical reactions mostly occurred in the sandstone for the first 2000 a with some precipitation at the ss/sh interface. From 2000 to 4000 a, ankerite, dolomite and illite began replacing Mg–Fe chlorite at the sandstone/shale interface. Transformation of chlorite to ankerite is the dominant reaction occluding the shale porosity in most simulations: from 4000 to 7500 a, this carbonation seals the reservoir and terminates reaction. Overall, the carbonates (calcite, ankerite, dolomite), chlorite and goethite all remain close to local chemical equilibrium with brine. Quartz is almost inert from the point of its dissolution/precipitation. However, the rate of quartz reaction controls the long-term decline in aqueous silica activity and its evolution toward equilibrium. The reactions of feldspars and clays depend strongly on their reaction rate constants (microcline is closer to local equilibrium than albite). The timing of porosity occlusion mostly therefore depends on the kinetic constants of kaolinite and illite. For example, an increase in the kaolinite kinetic constant by 0.25 logarithmic units hastened porosity closure by 4300 a. The earliest simulated closure of porosity occurred at approximately 108 a for simulations designed as sensitivity tests for the rate constants. These simulations also emphasize that the rate of CO 2 immobilization as aqueous bicarbonate (solubility trapping) or as carbonate minerals (mineral trapping) in sandstone reservoirs depends upon reaction kinetics – but the relative fraction of each trapped CO 2 species only depends upon the initial chemical composition of the host sandstone. For example, at the point of porosity occlusion the fraction of bicarbonate remaining in solution depends upon the initial Na and K content in the host rock but the fraction of carbonate mineralization depends only on the Ca, Mg, Fe content. Since ankerite is the dominant mineral that occludes porosity, the dissolved concentration of ferrous iron is also an important parameter. Finally, future efforts should focus on cross-comparisons and ground-truthing of simulations made for standard case studies as well as laboratory measurements of the reactivities of clay minerals.

58 GEOSCIENCES↗

Site-Specific Preparation of Intact Solid–Liquid Interfaces by Label-Free In Situ Localization and Cryo-Focused Ion Beam Lift-Out

Abstract Scanning transmission electron microscopy (STEM) allows atomic scale characterization of solid–solid interfaces, but has seen limited applications to solid–liquid interfaces due to the volatility of liquids in the microscope vacuum. Although cryo-electron microscopy is routinely used to characterize hydrated samples stabilized by rapid freezing, sample thinning is required to access the internal interfaces of thicker specimens. Here, we adapt cryo-focused ion beam (FIB) “lift-out,” a technique recently developed for biological specimens, to prepare intact internal solid–liquid interfaces for high-resolution structural and chemical analysis by cryo-STEM. To guide the milling process we introduce a label-free in situ method of localizing subsurface structures in suitable materials by energy dispersive X-ray spectroscopy (EDX). Monte Carlo simulations are performed to evaluate the depth-probing capability of the technique, and show good qualitative agreement with experiment. We also detail procedures to produce homogeneously thin lamellae, which enable nanoscale structural, elemental, and chemical analysis of intact solid–liquid interfaces by analytical cryo-STEM. This work demonstrates the potential of cryo-FIB lift-out and cryo-STEM for understanding physical and chemical processes at solid–liquid interfaces.

36 MATERIALS SCIENCE↗

In Situ Velocity‐Strain Sensitivity Near the San Jacinto Fault Zone Analyzed Through Train Tremors

Abstract We utilize train tremors as P‐wave seismic sources to investigate velocity‐strain sensitivity near the San Jacinto Fault Zone. A dense nodal array deployed at the Piñon Flat Observatory is used to detect and identify repeating train energy emitted from a railway in the Coachella valley. We construct P‐wave correlation functions across the fault zone and estimate the spatially averaged dt / t versus strain sensitivity to be 6.25 × 10 4 . Through numerical simulations, we explore how the sensitivity decays exponentially with depth. The optimal solution reveals a subsurface sensitivity of 1.2 × 10 5 and a depth decay rate of 0.05 km −1 . This sensitivity aligns with previous findings but is toward the higher end, likely due to the fractured fault‐zone rocks. The depth decay rate, previously unreported, is notably smaller than assumed in empirical models. This raises the necessity of further investigations of this parameter, which is crucial to study stress and velocity variations at seismogenic depth.

Geology↗

Advanced Terrestrial Simulator (ATS) evaluation dataset at 7 catchments across the continental United States

This dataset comprises of the input files and other files required for Advanced Terrestrial Simulator (ATS) simulations at 7 catchments across the continental United States. ATS is an integrated surface-subsurface hydrology model. We include Jupyter notebooks (within scripts folder) for individual catchments showing information (including data sources, river network, soil, geology, landuse types etc.) on preparing the machine readable input files. ATS observation output files are provided in the output folder. Figures and analyses (.xlsx sheets) are also provided. The catchments include, Taylor River Upstream (Colorado); (b) Cossatot River (Arkansas); (c) Panther Creek (Alabama); (d) Little Tennessee River (North Carolina and Georgia); (e) Mayo River (Virginia); (f) Flat Brook (New Jersey); (g) Neversink River headwaters (New York). Readme files are provided inside the directories providing more details. Files types include: .xml, .h5, .xlsx, .png, .ipynb, .py, .nc, .txt. All of the files types can be accessed by open source software, details on software requirements are following: .xml (any text editors including notepad and textedit), .h5 (in python using hdf libraries), .xlsx (WPS Office Spreadsheets, OpenOffice Calc, LibreOffice Calc, Microsoft Office etc.), .png (any image viewer), .ipynb (Jupyter notebook), .py (any text editors including notepad and textedit), .nc (using python or other open source software).

54 ENVIRONMENTAL SCIENCES↗

Machine-Learning of Nonlocal Kernels for Anomalous Subsurface Transport from Breakthrough Curves

Anomalous behavior is ubiquitous in subsurface solute transport due to the presence of high degrees of heterogeneity at different scales in the media. Although fractional models have been extensively used to describe the anomalous transport in various subsurface applications, their application is hindered by computational challenges. Simpler nonlocal models characterized by integrable kernels and finite interaction length represent a computationally feasible alternative to fractional models; yet, the informed choice of their kernel functions still remains an open problem. We propose a general data-driven framework for the discovery of optimal kernels on the basis of very small and sparse data sets in the context of anomalous subsurface transport. Using spatially sparse breakthrough curves recovered from fine-scale particle-density simulations, we learn the best coarse-scale nonlocal model using a nonlocal operator regression technique. Predictions of the breakthrough curves obtained using the optimal nonlocal model show good agreement with fine-scale simulation results even at locations and time intervals different from the ones used to train the kernel, confirming the excellent generalization properties of the proposed algorithm. A comparison with trained classical models and with black-box deep neural networks confirms the superiority of the predictive capability of the proposed model.

97 MATHEMATICS AND COMPUTING↗