Computational Box Research Articles

The effect of external force on the crack propagation of aluminum nanoplate using molecular dynamics approach: Insights into the fracture mechanisms of metallic nanomaterials under external loading condition

It is crucial to comprehend how external forces (EFs) affect crack propagation (CP) in aluminum (Al) nanoplates to develop and create nanomaterials with enhanced mechanical characteristics. The creation of novel materials for a variety of uses, such as the aerospace, electronics, and energy sectors, may benefit from this expertise. Additionally, insights into the fracture mechanisms of nanomaterials can aid in designing more reliable and durable structures at the nanoscale. This study utilized computer models to investigate the effect of EFs on fractures in Al nanoplates. The results suggest that an EF can significantly alter CP within nanoplates. The findings provide insights into the fracture mechanisms of metallic nanomaterials under external loading conditions. Simulation results in current research showed the physical stability of modeled Al nanoplates at T=300 K as the initial temperature. Numerically, the total energy (TE) of pristine nanoplate converged to −34762.953 eV after thermodynamic equilibrium detection inside the computational box. Furthermore, the simulation results show that EF caused the crack growth procedure intensity to increase. In the present study, the crack length value increased to 33.902 Å between our modeled samples. This result led to the conclusion that in real-world applications, it is important to consider the effect of EFs on the development of cracks within Al nanoplates.

Development of an ensemble CNN model with explainable AI for the classification of gastrointestinal cancer.

The implementation of AI assisted cancer detection systems in clinical environments has faced numerous hurdles, mainly because of the restricted explainability of their elemental mechanisms, even though such detection systems have proven to be highly effective. Medical practitioners are skeptical about adopting AI assisted diagnoses as due to the latter's inability to be transparent about decision making processes. In this respect, explainable artificial intelligence (XAI) has emerged to provide explanations for model predictions, thereby overcoming the computational black box problem associated with AI systems. In this particular research, the focal point has been the exploration of the Shapley additive explanations (SHAP) and local interpretable model-agnostic explanations (LIME) approaches which enable model prediction explanations. This study used an ensemble model consisting of three convolutional neural networks(CNN): InceptionV3, InceptionResNetV2 and VGG16, which was based on averaging techniques and by combining their respective predictions. These models were trained on the Kvasir dataset, which consists of pathological findings related to gastrointestinal cancer. An accuracy of 96.89% and F1-scores of 96.877% were attained by our ensemble model. Following the training of the ensemble model, we employed SHAP and LIME to analyze images from the three classes, aiming to provide explanations regarding the deterministic features influencing the model's predictions. The results obtained from this analysis demonstrated a positive and encouraging advancement in the exploration of XAI approaches, specifically in the context of gastrointestinal cancer detection within the healthcare domain.

Persistent mixed‐phase states in adiabatic cloud parcels under idealised conditions

AbstractBy assuming that liquid droplets and ice crystals within a computational grid box grow under the same conditions, mean‐field representations of mixed‐phase clouds in numerical models favour a quick cloud glaciation driven by the Wegener–Bergeron–Findeisen process. Consequently, maintenance of mixed‐phase states under the mean‐field approximation is conditioned to external dynamical forcing, such as sufficiently strong updraughts. In this work we go beyond the mean‐field representation and investigate the maintenance of mixed‐phase states in adiabatic (non‐entraining) cloud volumes by accounting for local variability in a particle's growth conditions in the turbulent cloud environment. This is done by using a Lagrangian microphysical scheme, where temperature and vapour mixing ratio are stochastic attributes attached to each cloud particle. Different dynamic scenarios show that microphysical variability and parametrised turbulence effects may significantly reduce cloud glaciation rates, resulting in much more resilient mixed‐phase states in idealised parcels containing non‐sedimenting and non‐aggregating cloud particles. We have stated a more refined criterion for the Wegener–Bergeron–Findeisen process activity in the bulk of a mixed‐phase cloud parcel (or computational grid box). This criterion is stated in terms of the conditional average supersaturations that are experienced by specific cloud particle types (liquid or ice), and not in terms of unconditional averages corresponding to parcel‐ or grid‐mean values of supersaturations. Formulation of a relation between conditional and unconditional average supersaturations poses an interesting closure problem in mixed‐phase cloud microphysics. Our stochastic microphysical model provides a Lagrangian closure to this problem and gives insights towards the development of a prognostic stochastic subgrid‐scale scheme for condensation/deposition in numerical models of mixed‐phase clouds.

A tale of a changing basin - a transient model of the 7.17 event leading to the Messinian Salinity Crisis

Before the Messinian Salinity Crisis (MSC) left its imprint on the sediment record of the Mediterranean Sea in the form of evaporites, the basin had already undergone significant changes. At 7.17 Ma, a drop in δ13C values, as well as a basin-wide shift in the abundance of benthic foraminifers, already attest to a sudden change in the Mediterranean conditions.This event coincides with an increase in the amplitude of the insolation curve. It thus stands to question whether a change in the freshwater budget or a change in the connection between the Mediterranean Sea and the Atlantic was the driver for this event. Answering this question would not only help to understand the event itself, but might also help to decipher the early dynamics of the MSC.With a computational box model, we investigate the response of the Mediterranean Sea to a varying freshwater budget for a wide range of restriction. The results then let us define scenarios in which we analyse how a gradually changing restriction would express itself in the basin dynamics.We find that the change in the freshwater budget alone cannot explain the changes that are attributed with the 7.2 event, but coupled with an increase in restriction most differences can be accounted for. Our results also show that a gradual change in restriction can provoke a non-linear response in the behaviour of the basin, which can appear abrupt when happening on a short enough timescale. Such a change would also enhance the influence of said changes in the freshwater budget.This tells us that the processes that most likely triggered the Messinian Salinity Crisis started much earlier and incrementally increased the restriction of the Mediterranean Sea.

A numerical study of the effect of variable heat flux on the stability and thermal behavior of SARS-COV-2 structure: A molecular dynamics approach

One of the common methods is the molecular dynamics simulation which models the behavior of atoms and molecules. This paper used the molecular dynamics technique to simulate the behavior of SARS-COV-2 virus under variable heat flux conditions. By doing so, it can be observed how the virus structure responded to the changes in external heat flux and how this affected its stability. This paper studied the effect of external heat flux with different amplitudes of 0.1, 0.2, 0.3, and 0.5 W/m2 on the stability of SARS in an aqueous medium. The present study showed that the implementation of external heat flux to modeled samples significantly affected their physical stability. Numerically, the mean square displacement of system decreased to 0.634 nm2 by increasing the heat flux inside the computational box. This atomic evolution predicted the stability of target structure increased by heat flux implementation to samples. Physically, this behavior arose from increasing attraction force among various particles inside the SARS-COV-2 structure in the presence of external heat flux. So, we expect this atomic evolution in treatment method design in clinical cases.

The effect of initial temperature on the mechanical properties of paraffin-reinforced silica aerogel using molecular dynamics simulation: Application in semi-transparent roofs, electric vehicles and building insulation

Aerogels are formed from a lattice of interconnected nanoparticles with a nanometer space between them. These nanostructured materials, with very low density and high specific surfaces, show exceptional properties, so their applications in various fields are expanding. In the present study, the mechanical behavior of paraffin-reinforced silica aerogel was examined by the change in the stress-strain curve, ultimate strength (US), and Young's modulus (YM) at different initial temperatures of 300, 310, 320, 330, and 350 K by MD and LAMMPS. It is revealed that rising the initial temperature can affect the silica aerogel's mechanical properties and impair the structure's mechanical strength. By increasing the initial temperature from 300 to 350 K, the US and YM modules declined from 410.45 to 1073.13 to 351.36–661.4 MPa, respectively. Physically, this mechanical performance results from increased atomic oscillations inside the computational box. Therefore, it is concluded that the effect of temperature on silica's mechanical performance is important and used in various industries.

Advantages of Multidimensional Biasing in Accelerated Dynamics: Application to the Calculation of the Acid pKa for Acetic Acid.

The use of accelerated sampling methods such as metadynamics has shown a significant advantage in calculations that involve infrequent events, which would otherwise require sampling a prohibitive number of configurations to determine the difference in free energies between two or more chemically distinct states such as in the calculation of acid dissociation constants Ka. In this case, the most common method is to bias the system via a single collective variable (CV) representing the coordination number of the proton donor group, which yields results in reasonable agreement with experiments. Here we study the deprotonation of acetic acid using the reactive force field ReaxFF and observe a significant dependence of Ka on the simulation box size when biasing only the coordination number CV, which is due to incomplete sampling of the deprotonated state for small simulation systems and inefficient sampling for larger ones. Incorporating a second CV representing the distance between the H3O+ cation and the acetate anion results in substantially more efficient sampling, both accelerating the dynamics and virtually eliminating the computational box size dependence.

Boundary conditions for molecular simulations of isolated elastic defects. Case study: The ⟨111⟩ screw dislocation in bcc W

A new set of boundary conditions is proposed for molecular simulations of isolated elastic defects such as dislocations and cracks. The case study of the ⟨111⟩ screw dislocation in body centered cubic (bcc) tungsten, modeled via a phenomenological, n-body cohesion functional, serves validating the new boundary conditions by computing structural properties of this defect and comparing these with results from the literature. Lowest energy configurations of the dislocated crystal have been obtained by molecular statics incorporating the new boundary conditions. The associated displacement and energy landscapes reveal conformal to the predictions of the elastic theory for a screw dislocation embedded in an infinitely extended crystal. In particular, no energy gradients and positional mismatch of atoms are found at the terminations of the computational box, validating thereby the new boundary conditions. Furthermore, it is shown that the structure, the spatial extension, and the excess energy of the two possible core polarizations of this dislocation compare consistently with existing findings for this and other bcc metals. Close to the dislocation line, energy minimization triggers the emergence of anelastic edge displacements extending over distances unexpectedly much larger than the dislocation core radius. Therefore, the conclusion is reached that in molecular simulations, the transverse to the dislocation line dimensions of the atomistic model should be taken considerably larger than it is accustomed. Perspectives opened by the present work are briefly discussed.

AI framework with computational box counting and Integer programming removes quantization error in fractal dimension analysis of optical images

Because many Chemical Engineering phenomena and processes involve distinctive shapes and structures, fractal dimension (FD) analysis is of great interest to the Chemical Engineering community due to its utility in providing a statistical index of shape complexity, which not only quantifies image structure but also helps explain functional properties. Past box counting (BC) methods for estimation of FDs are inconsistent due to quantization error (QE) introduced from image rotation and translation. In this work, we propose a systematic and automatic artificial intelligence (AI) framework that consistently estimates FDs of relevance to different fields of Chemical Engineering without QE by integrating image preprocessing, set-covering optimization-based BC, and regression analysis. As a result of the deterministic optimization technique, FD estimations remain consistent for all images regardless of image rotation and translation. The results of image datasets obtained from fields such as interfacial science, biomedical engineering, human anatomy, among others, demonstrate efficient box size specification to maximize regression effectiveness and FD calculation accuracy for a variety of images. Overall, the proposed AI framework offers a new means of estimating FD accurately and efficiently for optical images of interest to the Chemical Engineering community.

Multiscale analysis of some shear layers in a fully developed turbulent channel flow

Multiscale edge detection-wavelet analysis is applied to the velocity field emanating from direct numerical simulations of a fully developed turbulent channel flow. The direct numerical simulations are realized in large computational domains at four different Karman numbers up to 1100. The shear layers (SL) related to the streamwise and spanwise velocity fields involving streamwise and spanwise gradients are analyzed scale-by-scale, over the homogeneous planes. The SL involving the spanwise gradient, in particular the wall vorticity layers surrounding the quasi-streamwise vortices (QSV) are organized into long streamwise streaks, whose spanwise spacing and thickness are determined versus the wavelet-wavenumber. The shear layers related to the streamwise gradient are further analyzed by using the Rice decomposition of the wavelet coefficients into a local phase and a local amplitude. Large zones of constant phase with slowly varying local amplitude have been identified at different scales. It is found that the constant phase-locked regions coincide with the wakes of the buffer layer QSV. The arrival of the quasi-streamwise vortices cause phase jumps and large intermittency in the local amplitude. The analogy between these peculiarities and the imperfect chaos synchronization phenomena together with the consequences in term of wall turbulence control are discussed. These results confirm globally our previous investigation realized through a low Reynolds number channel flow DNS in a small computational box and clarify some points that could not then elucidated.

Open-channel flow over evolving subaqueous ripples

We have numerically investigated the turbulent flow and sediment grain motion in an open-channel flow configuration over a subaqueous sediment bed featuring two-dimensional transverse ripples at moderate Reynolds number and super-critical Shields number values. The simulation data, which were generated by means of particle-resolved direct numerical simulation, are the same as in our previous work (Kidanemariam & Uhlmann,J. Fluid Mech., vol. 818, 2017, pp. 716–743). By carefully choosing the computational box sizes, we were able to accommodate single ripple units which form over an initially flat sediment bed at a wavelength equal to the domain length. The ripples then evolve into their asymmetric shape relatively quickly and eventually migrate downstream steadily while maintaining their shape and size. In the present study, using a ripple-conditioned phase-averaging procedure, we are able to obtain novel insights into the evolution of the turbulent flow and particle motion over the bedforms, in particular the spatial structure of the basal shear stress and its relation to the particle flow rate. Our analysis confirms that the boundary shear-stress maximum is located upstream of the ripple crest, while the particle flow rate is essentially in phase with the ripple topology, with an average phase difference between the two in the range of 18–19 particle diameters for the considered parameter values. We were further able to confirm the link between the sediment flux relaxation behaviour and the observed shear-stress/geometry lag, by direct evaluation of the saturation length scale.

Spanwise-coherent hydrodynamic waves around flat plates and airfoils

We investigate spanwise-coherent structures in the turbulent flow around airfoils, motivated by their connection with trailing-edge noise. We analyse well-resolved large-eddy simulations (LES) of the flow around NACA 0012 and NACA 4412 airfoils, both at a Reynolds number of 400 000 based on the chord length. Spectral proper orthogonal decomposition performed on the data reveals that the most energetic coherent structures are hydrodynamic waves, extending over the turbulent boundary layers around the airfoils with significant amplitudes near the trailing edge. Resolvent analysis was used to model such structures, using the mean field as a base flow. We then focus on evaluating the dependence of such structures on the domain size, to ensure that they are not an artefact of periodic boundary conditions in small computational boxes. To this end, we performed incompressible LES of a zero-pressure-gradient turbulent boundary layer, for three different spanwise sizes, with the momentum-thickness Reynolds number matching those near the airfoils trailing edge. The same coherent hydrodynamic waves were observed for the three domains. Such waves are accurately modelled as the most amplified flow response from resolvent analysis. The signature of such wide structures is seen in non-premultiplied spanwise wavenumber spectra, which collapse for the three computational domains. These results suggest that the spanwise-elongated structures are not domain-size dependent for the studied simulations, indicating thus the presence of very wide structures in wall-bounded turbulent flows.

A numerical study on elastic properties of low-density two-dimensional networks of crosslinked long fibers

Fiber-based materials are prevalent in nature and in engineering applications. Microscopically, these materials resemble a discrete assembly of crosslinked or entangled fibers. To understand the relationship between the effective mechanical properties and the underlying microstructures, we consider a variety of periodic and random two-dimensional (2D) networks of crosslinked long fibers. The linearly elastic properties of periodic 2D networks (e.g., square, triangular and Kagome) are well understood. However, for low-density networks, cooperative buckling of the fiber segments can take place at small strains, leading to nonlinear, anisotropic elastic behaviors. A transition from stretch to bending and then back to stretch dominated deformation is predicted for the Kagome and triangular networks. For random 2D networks, the elastic behaviors are different. Under uniaxial tension, the stress–strain behavior is statistically isotropic and slightly nonlinear, dominated by stretch of the fibers aligned closely to the loading direction. Meanwhile, stochastic buckling occurs continuously in the random networks, leading to significant lateral contraction. Consequently, while the effective Young’s modulus follows a nearly linear scaling with respect to the relative density, the effective Poisson’s ratio exhibits a transition from stretch to bending dominated mode as the relative density decreases. A statistical analysis is performed to estimate the relative errors of the effective properties that depend on both the computational box size and the number of random realizations. The comparison between the periodic and random 2D networks highlights the profound effects of the network topology on the effective elastic properties.

Social choice using moral metrics

PurposeThis paper aims to develop a geometry of moral systems. Existing social choice mechanisms predominantly employ simple structures, such as rankings. A mathematical metric among moral systems allows us to represent complex sets of views in a multidimensional geometry. Such a metric can serve to diagnose structural issues, test existing mechanisms of social choice or engender new mechanisms. It also may be used to replace active social choice mechanisms with information-based passive ones, shifting the operational burden.Design/methodology/approachUnder reasonable assumptions, moral systems correspond to computational black boxes, which can be represented by conditional probability distributions of responses to situations. In the presence of a probability distribution over situations and a metric among responses, codifying our intuition, we can derive a sensible metric among moral systems.FindingsWithin the developed framework, the author offers a set of well-behaved candidate metrics that may be employed in real applications. The author also proposes a variety of practical applications to social choice, both diagnostic and generative.Originality/valueThe proffered framework, derived metrics and proposed applications to social choice represent a new paradigm and offer potential improvements and alternatives to existing social choice mechanisms. They also can serve as the staging point for research in a number of directions.

Unravelling Neural Coding of Dynamic Natural Visual Scenes via Convolutional Recurrent Neural Networks

Retina, providing and preprocessing almost all the visual information for brain, is a much sophisticated neural system than we think. In neural coding, the central goal is to finding the relationship between a cause (e.g. a visual stimulus) and the effect (i.e., the neural response). Neural system identification aims to reveal the computational black boxes by learning the stimulus-response relationship. Traditional models of retinal system identification analysis the neural respond to artificial stimulus by models consisted of predefined components. The design of the model is limited by prior knowledge, and the artificial stimulus is too simple to be compared with the stimulus actually processed by the retina. As the most successful prediction models, deep neural networks, have demonstrated powerful computational ability in neural coding. In this study, to fill in the gap of explainable model that reveals how population of neurons work together to encode the larger field of dynamic natural scenes, we used a deep learning model to identify the computational elements of the retinal neural system that contribute to learning of the dynamics of natural visual scenes. Specifically, the proposed model can separate intricate spatiotemporal patterns by leveraging building blocks of the neural network model. By using physiological data with sequential natural scene inputs, we verify that the recurrent connection plays a key role in encoding complex dynamic visual scenes while learning biological computational underpinnings of the retina circuit. Our model subsequently learns both the shapes and location of the spatiotemporal receptive fields of ganglion cells. Moreover, we propose a method for evaluating the importance of the kernels learned by the network model in encoding dynamic visual scenes, which allows us to disentangle the dynamic computational structure of the retina. These results provide new insights into understanding the encoding mechanisms of retinal neurons and promote model design in machine learning for computational vision involving complex dynamic visual scenes.

Structure interactions in a reduced-order model for wall-bounded turbulence

New reduced-order models (ROMs) are derived for sinusoidal shear flow (also known as Waleffe flow) and plane Couette flow in small periodic domains. A first derivation for Waleffe flow exploits Fourier modes that form a natural orthonormal basis for the problem. A ROM for such basis is obtained by a Galerkin projection of the Navier-Stokes equation. A large basis was reduced to 12 modes that contribute significantly in maintaining chaotic, turbulent dynamics. A key difference from earlier ROMs is the inclusion of two roll-streak structures, with spanwise wavelengths equal to $L_z$ and $L_z/2$, where $L_z$ is the spanwise length of the computational box. The resulting system was adapted to Couette flow by rewriting the Galerkin system for the same 12 modes, modified so as to satisfy no-slip conditions on the walls. The resulting dynamical systems lead to turbulence with finite lifetimes, in agreement with earlier ROMs and simulations in small domains. However, the present models display lifetimes that are much longer than in earlier ROMs, with differences of more than an order of magnitude. The Couette-flow model is compared to results of direct numerical simulation (DNS), with statistics displaying fair agreement. The inclusion of the $L_z$ and $L_z/2$ lengthscales is seen to be a key feature for longer turbulence lifetimes: neglecting any of the roll modes, or their non-linear interaction, leads to drastic reductions of turbulence lifetimes. The present ROMs thus highlight some of the dominant nonlinear interactions that are relevant in maintaining turbulence for long lifetimes.

Dynamics of shear-layer coherent structures in a forced wall-bounded flow

Abstract

Interleaving physics- and data-driven models for power system transient dynamics

The paper explores interleaved and coordinated refinement of physics- and data-driven models in describing transient phenomena in large-scale power systems. We develop and study an integrated analytical and computational data-driven gray box environment needed to achieve this aim. Main ingredients include computational differential geometry-based model reduction, optimization-based compressed sensing, and a finite approximation of the Koopman operator. The proposed two-step procedure (the model reduction by differential geometric (information geometry) tools, and data refinement by the compressed sensing and Koopman theory based dynamics prediction) is illustrated on the multi-machine benchmark example of IEEE 14-bus system with renewable sources, where the results are shown for doubly-fed induction generator (DFIG) with local measurements in the connection point. The algorithm is directly applicable to identification of other dynamic components (for example, dynamic loads).

A lattice Boltzmann study on Brownian diffusion and friction of a particle in a confined multicomponent fluid

We study the diffusivity of a small particle immersed in a square box filled with a non-ideal multicomponent fluid in the presence of thermal fluctuations. Our approach is based on the numerical integration of fluctuating lattice Boltzmann models (LBM) for multicomponent mixtures. At changing the wettability on the particle's surface, we measure the mean square displacement (MSD) and compare with the prediction of the Stokes-Einstein theory. Two main set-ups are tested, involving periodic boundary conditions and wall boundary conditions realized on the computational box. We find that full periodic boundary conditions give rise to random advection after millions of lattice Boltzmann time steps, while this effect is mitigated in the presence of wall boundary conditions. The matching with the Stokes-Einstein relation is therefore guaranteed when we use the appropriate frictional properties measured in the presence of confinement. Our results will help the exploration of nanoscale applications with multicomponent fluids using LBM in the presence of thermal fluctuations.

On rising magnetic flux tube and formation of sunspots in a deep domain

ABSTRACT We investigate the rising flux tube and the formation of sunspots in an unprecedentedly deep computational domain that covers the whole convection zone with a radiative magnetohydrodynamics simulation. Previous calculations had shallow computational boxes (<30 Mm) and convection zones at a depth of 200 Mm. By using our new numerical code Radition and RSST for Deep Dynamics(r2d2), we succeed in covering the whole convection zone and reproduce the formation of the sunspot from a simple horizontal flux tube because of the turbulent thermal convection. The main findings are as follows. (1) The rising speed of the flux tube is larger than the upward convection velocity because of the low density caused by the magnetic pressure and the suppression of the mixing. (2) The rising speed of the flux tube exceeds 250 m s−1 at a depth of 18 Mm, while we do not see any clear evidence of the divergent flow 3 h before the emergence at the solar surface. (3) Initially, the root of the flux tube is filled with the downflows, and then the upflow fills the centre of the flux tube during the formation of the sunspot. (4) The essential mechanisms for the formation of the sunspot are the coherent inflow and the turbulent transport. (5) The low-temperature region is extended to a depth of at least 40 Mm in the matured sunspot, with the high-temperature region in the centre of the flux tube. Some of the findings indicate the importance of the deep computational domain for the flux emergence simulations.