Energy Minimization Model Research Articles

The principle of compromise-in-competition: understanding mesoscale complexity of different levels

This perspective is not meant to be a complete summary of the full series of our studies over decades, rather it aims to look back across the entire process to outline: What inspired this work at its beginning? How did the mindset gradually evolve from a solely engineering study to the new concept of Mesoscience? And where is the research currently being focused and to be extended? This approach elucidates the central importance of correlating precise engineering problems to be resolved with associated missing links in the underlying, fundamental science while exploring a notable commonality, namely, the universality of Mesoscience. Inspired by the physical phenomenon of the coexistence and interaction between a gas-rich dilute phase and a solid-rich dense phase with different physical mechanisms dominating the behaviour for each phase in gas–solid fluidization, it was recognized that an underlying compromise-in-competition (CIC) exists between these two physically dominant mechanisms. This, then, is the CIC principle. We believe that this is the origin of inherent structural complexity in these and other systems. We have shown that this CIC principle can be formulated as a multi-objective optimization problem through the so-called energy-minimization multi-scale model, leading to a massive improvement in both the accuracy and scalability of computation and its wide application in both academia and industries. Furthermore, this approach was extended to formulate and understand other complex systems in many apparently disparate fields, indicating the wide applicability of the CIC principle and then resulting in the proposition and advancement of the concept of Mesoscience. Furthermore, within the main body of this perspective article, future directions for identifying, developing and applying the concept of Mesoscience will be prospected. This includes extending its possible generality and applications in many different disciplines and fields, even to analysing global challenging issues and promoting CIC-informed AI. In this way, many challenging problems in engineering—identified through their underlying mesoscale complexity—will be closely correlated with the development of future fundamental science. Through this approach, we hope to illustrate that fundamental research can now tackle the intrinsic complexity existing in a multitude of engineering problems. Through the concept of Mesoscience, we aim to explore the possible commonality from complexity. The deduction of commonality from diversity is possible in resolving global challenges, shifting paradigms in science and filling in the existing gaps at the mesoscales of different levels of our knowledge, though there are difficulties and uncertainty coming from diversity.

Gas-liquid flow regimes in a novel rocking and rolling flow loop

Predicting two-phase flow pattern characteristics and flow transition is fundamental to address several industrial problems, e.g. for the oil and gas industry. In order to fulfil the upscaling from the laboratory to the industrial scale, resource-efficient flow testing facilities which closely replicate industrial flow characteristics are needed. To address this, we introduce a new experimental device named as the Rocking and Rolling Ring Flow Loop (3RFL), which is size, cost, and time-efficient. In the present work, we employ the 3RFL apparatus at atmospheric pressure and temperature, with air and water as working fluids. However, the ultimate goal of our work is to build an experimental set-up capable of capturing the under-pressure reactive multiphase flow (gas-liquid–solid) dynamics typical of flow assurance in oil production and transportation. At the moment, the 3RFL can induce different flow regimes by adjusting the system control parameters such as rocking angle, rocking rate, and liquid volume fraction, all without requiring a pump. We observe three flow regimes, and analyse the impact of control parameters on their emergence. Our findings reveal that flow regime transitions are influenced by the competition between gravitational and centrifugal forces, which arise due to the curvature of the tube. Among three employed modelling strategies, namely mechanistic modelling, total energy minimization and a combined approach, we find that the total energy minimization model best compares to the experimental liquid height.

Non-orthogonal two-step annealing method for linearized magnetic tunnel junction sensors

The orthogonal two-step annealing process is an effective strategy to linearize the response of magnetic tunnel junctions for magnetic field sensors. However, the response after the orthogonal annealing is inevitably modulated by the Neel effect from the reference layer, which results in an unexpected shift of the linear interval and a disappointing sensitivity deterioration in the weak field. Here, a non-orthogonal two-step annealing method is proposed to suppress the shift by compensating for the Neel coupling field. Experimental results show that the curve shift of junctions annealed in the non-orthogonal direction of 120° is 47.6% lower than that in the orthogonal one, with a significant sensitivity promotion in the weak field and little hysteresis increment. A simple energy minimization model is introduced to explain the results. Based on the model, the suppression of the curve shift is fulfilled with the effective field compensation for the Neel coupling field, modulated by the non-orthogonal annealing. Finally, Wheatstone bridge devices are constructed, and the bridge with non-orthogonally annealed junctions is found to have an increased sensitivity of 46.8% in the major loop along the sensing axis. Additionally, the non-orthogonal annealing method is also effective in suppressing the cross sensitivity, which is important for further application to three-axis magnetic sensors.

Learning Graph Attentions via Replicator Dynamics.

Graph Attention (GA) which aims to learn the attention coefficients for graph edges has achieved impressive performance in GNNs on many graph learning tasks. However, existing GAs are usually learned based on edges' (or connected nodes') features which fail to fully capture the rich structural information of edges. Some recent research attempts to incorporate the structural information into GA learning but how to fully exploit them in GA learning is still a challenging problem. To address this challenge, in this work, we propose to leverage a new Replicator Dynamics model for graph attention learning, termed Graph Replicator Attention (GRA). The core of GRA is our derivation of replicator dynamics based sparse attention diffusion which can explicitly learn context-aware and sparse preserved graph attentions via a simple self-supervised way. Moreover, GRA can be theoretically explained from an energy minimization model. This provides a more theoretical justification for the proposed GRA method. Experiments on several graph learning tasks demonstrate the effectiveness and advantages of the proposed GRA method on ten benchmark datasets.

Dynamic energy-minimization multi-scale heterogeneous continuum evolution model for gas-solid fluidization

Traditional gas–solid continuum model is still computationally inaccurate and physically unreasonable, due to a homogeneous assumption that particles are uniformly dispersed in a computational grid. Therefore, its application to simulate industrial gas–solid fluidization needs further improvement. Here, a novel heterogeneous continuum model, named dynamic energy-minimization multi-scale (EMMS) evolution model, was developed for accurate and fast simulations of gas–solid fluidization systems. In this model, the evolution of gas–solid heterogeneous multiscale structure in each computational grid was directly resolved by EMMS model, which was then coupled with multiphase mixture model to dynamically calculate phase holdup and flow field. Both S-shaped axial voidage profiles and spatial distribution of solid phase with the presence of a core-annulus structure were successfully captured. Compared with two-fluid model, dynamic EMMS evolution model showed great advantages in various aspects, including (a) model accuracy: predicted “S” type distribution of bed voidage was closer to experimental data; (b) stability: a rather larger time step of 0.1 s could be used; and (c) computational cost: dynamic EMMS evolution model was 245 times faster than two-fluid model. Based on dynamic EMMS evolution model, the complex multiscale hydrodynamics can be well simulated and understood, and will help the design and scale-up of large-scale gas–solid fluidized bed reactors.

Reciprocal intra- and extra-cellular polarity enables deep mechanosensing through layered matrices

Adherent cells migrate on layered tissue interfaces to drive morphogenesis, wound healing, and tumor invasion. Although stiffer surfaces are known to enhance cell migration, it remains unclear whether cells sense basal stiff environments buried under softer, fibrous matrix. Using layered collagen-polyacrylamide gel systems, we unveil a migration phenotype driven by cell-matrix polarity. Here, cancer (but not normal) cells with stiff base matrix generate stable protrusions, faster migration, and greater collagen deformation because of "depth mechanosensing" through the top collagen layer. Cancer cell protrusions with front-rear polarity produce polarized collagen stiffening and deformations. Disruption of either extracellular or intracellular polarity via collagen crosslinking, laser ablation, or Arp2/3 inhibition independently abrogates depth-mechanosensitive migration of cancer cells. Our experimental findings, validated by lattice-based energy minimization modeling, present a cell migration mechanism whereby polarized cellular protrusions and contractility are reciprocated by mechanical extracellular polarity, culminating in a cell-type-dependent ability to mechanosense through matrix layers.

Analysis of the interaction of a non-canonical twin half-site of Cyclic AMP-Response Element (CRE) with CRE-binding protein

Differential regulation of a gene having either canonical or non-canonical cyclic AMP response element (CRE) in its promoter is primarily accomplished by its interactions with CREB (cAMP-response element binding protein). The present study aims to delineate the mechanism of the CREB-CRE interactions at the Oncostatin-M (osm) promoter by in vitro and in silico approaches. The non-canonical CREosm consists of two half-CREs separated by a short intervening sequence of 9 base pairs. In this study, in vitro binding assays revealed that out of the two CRE half-sites, the right half-CRE was indispensable for binding of CREB, while the left sequence showed weaker binding ability and specificity. Genome-wide modeling and high throughput free energy calculations for the energy-minimized models containing CREB-CREosm revealed that there was no difference in the binding of CREB to the right half of CREosm site when compared to the entire CREosm. These results were in accordance with the in vitro studies, confirming the indispensable role of the right half-CREosm site in stable complex formation with the CREB protein. Additionally, conversion of the right half-CREosm site to a canonical CRE palindrome showed stronger CREB binding, irrespective of the presence or absence of the left CRE sequence. Thus, the present study establishes an interesting insight into the interaction of CREB with a CRE variant located at the far end of a TATA-less promoter of a cytokine-encoding gene, which in turn could be involved in the regulation of transcription under specific conditions.

Controlling aromatic helix dimerization in water by tuning charge repulsions.

Several helically folded aromatic oligoamides were designed and synthesized. The sequences were all water-soluble thanks to the charged side chains borne by the monomers. Replacing a few, sometimes only two, charged side chains by neutral methoxy groups was shown to trigger the formation of various aggregates which could be tentatively assigned to head-to-head stacked dimers of single helices, double helical duplexes and a quadruplex, none of which would form in organic solvent with organic-soluble analogues. The nature of the aggregates was supported by concentration and solvent dependent NMR studies, 1H DOSY experiments, mass spectrometry, and X-ray crystallography or energy-minimized models, as well as analogies with earlier studies. The hydrophobic effect appears to be the main driving force for aggregation but it can be finely modulated by the presence or absence of a small number of charges to an extent that had no precedent in aromatic foldamer architectures. These results will serve as a benchmark for future foldamer design in water.

Unification of free energy minimization, spatiotemporal energy, and dimension reduction models of V1 organization: Postnatal learning on an antenatal scaffold.

Developmental selection of neurons and synapses so as to maximize pulse synchrony has recently been used to explain antenatal cortical development. Consequences of the same selection process—an application of the Free Energy Principle—are here followed into the postnatal phase in V1, and the implications for cognitive function are considered. Structured inputs transformed via lag relay in superficial patch connections lead to the generation of circumferential synaptic connectivity superimposed upon the antenatal, radial, “like-to-like” connectivity surrounding each singularity. The spatiotemporal energy and dimension reduction models of cortical feature preferences are accounted for and unified within the expanded model, and relationships of orientation preference (OP), space frequency preference (SFP), and temporal frequency preference (TFP) are resolved. The emergent anatomy provides a basis for “active inference” that includes interpolative modification of synapses so as to anticipate future inputs, as well as learn directly from present stimuli. Neurodynamic properties are those of heteroclinic networks with coupled spatial eigenmodes.

The Characterization of a Novel Virus Discovered in the Yeast Pichia membranifaciens.

Mycoviruses are widely distributed across fungi, including the yeasts of the Saccharomycotina subphylum. This manuscript reports the first double-stranded RNA (dsRNA) virus isolated from Pichia membranifaciens. This novel virus has been named Pichia membranifaciens virus L-A (PmV-L-A) and is a member of the Totiviridae. PmV-L-A is 4579 bp in length, with RNA secondary structures similar to the packaging, replication, and frameshift signals of totiviruses that infect Saccharomycotina yeasts. PmV-L-A was found to be part of a monophyletic group within the I-A totiviruses, implying a shared ancestry between mycoviruses isolated from the Pichiaceae and Saccharomycetaceae yeasts. Energy-minimized AlphaFold2 molecular models of the PmV-L-A Gag protein revealed structural conservation with the Gag protein of Saccharomyces cerevisiae virus L-A (ScV-L-A). The predicted tertiary structure of the PmV-L-A Pol and other homologs provided a possible mechanism for totivirus RNA replication due to structural similarities with the RNA-dependent RNA polymerases of mammalian dsRNA viruses. Insights into the structure, function, and evolution of totiviruses gained from yeasts are essential because of their emerging role in animal disease and their parallels with mammalian viruses.

Flow characteristics simulation of spiral coil reactor used in the thermochemical energy storage system

According to environmental and energy issues, renewable energy has been vigorously promoted. Now solar power is widely used in many areas but it is limited by the weather conditions and cannot work continuously. Heat storage is a considerable solution for this problem and thermochemical energy storage is the most promising way because of its great energy density and stability. However, this technology is not mature enough to be applied to the industry. The reactor is an important component in the thermochemical energy storage system where the charging and discharging process happens. In this paper, a spiral coil is proposed and used as a reactor in the thermochemical energy storage system. The advantages of the spiral coil include simple structure, small volume, and so on. To investigate the flow characteristics, the simulation was carried out based on energy-minimization multi-scale model (EMMS) and Eulerian two-phase model. CaCO 3 particles were chosen as the reactants. Particle distribution was shown in the results. The gas initial velocity was set to 2 m·s −1 , 3 m·s −1 , and 4 m·s −1 . When the particles flowed in the coil, gravity, centrifugal force and drag force influenced their flow. With the Reynold numbers increasing, centrifugal and drag force got larger. Accumulation phenomenon existed in the coil and results showed with the gas velocity increasing, accumulation moved from the bottom to the outer wall of the coil. Besides, the accumulation phenomenon was stabilized when φ > 720°. Also due to the centrifugal force, a secondary flow formed, which means solid particles moved from the inside wall to the outside wall. This secondary flow could promote turbulence and mixing of particles and gas. In addition, when the particle volume fraction is reduced from 0.2 to 0.1, the accumulation at the bottom of the coil decreases, and the unevenness of the velocity distribution becomes larger.

Mathematical energy minimization model for joining boron nitride fullerene with several BN nanostructures

Nanoscale materials have gained considerable interest because of their special properties and wide range of applications. Many types of boron nitride at the nanoscale have been realized, including nanotubes, nanocones, fullerenes, tori, and graphene sheets. The connection of these structures at the nanoscale leads to merged structures that have enhanced features and applications. Modeling the joining between nanostructures has been adopted by different methods. Namely, carbon nanostructures have been joined by minimizing the elastic energy in symmetric configurations. In other words, the only considerable curvature in the elastic energy is the axial curvature. Accordingly, because it has nanoscale structures similar to those in carbon, BN can also be joined and connected by using this method. On the other hand, different methods have been proposed to consider the rotational curvature because it has a similar size. Based on that argument, the Willmore energy, which depends on both curvatures, has been minimized to join carbon nanostructures. This energy is used to identify the joining region, especially for a three-dimensional structure. In this paper, we expand the use of Willmore energy to cover the joining of boron nitride nanostructures. Therefore, because catenoids are absolute minimizers of this energy, pieces of catenoids can be used to connect nanostructures. In particular, we joined boron nitride fullerene to three other BN nanostructures: nanotube, fullerene, and torus. For now, there are no experimental or simulation data for comparison with the theoretical connecting structures predicted by this study, which is some justification for the suggested simple model shown in this research. Ultimately, various nanoscale BN structures might be connected by considering the same method, which may be considered in future work.

Induced forms of α2-macroglobulin neutralize heparin and direct oral anticoagulant effects

Alpha2-macroglobulin (α2M) is a physiological macromolecule that facilitates the clearance of many proteinases, cytokines and growth factors in human. Here, we explored the effect of induced forms of α2M on anticoagulant drugs. Gla-domainless factor Xa (GDFXa) and methylamine (MA)-induced α2M were prepared and characterized by electrophoresis, immunonephelometry, chromogenic, clot waveform and rotational thromboelastometry assays. Samples from healthy volunteers and anticoagulated patients were included. In vivo neutralization of anticoagulants was evaluated in C57Bl/6JRj mouse bleeding-model. Anticoagulant binding sites on induced α2M were depicted by computer-aided energy minimization modeling. GDFXa-induced α2M neutralized dabigatran and heparins in plasma and whole blood. In mice, a single IV dose of GDFXa-induced α2M following anticoagulant administration significantly reduced blood loss and bleeding time. Being far easier to prepare, we investigated the efficacy of MA-induced α2M. It neutralized rivaroxaban, apixaban, dabigatran and heparins in spiked samples in a concentration-dependent manner and in samples from treated patients. Molecular docking analysis evidenced the ability of MA-induced α2M to bind non-covalently these compounds via some deeply buried binding sites. Induced forms of α2M have the potential to neutralize direct oral anticoagulants and heparins, and might be developed as a universal antidote in case of major bleeding or urgent surgery.

In silico structural elucidation of the rabies RNA-dependent RNA polymerase (RdRp) toward the identification of potential rabies virus inhibitors.

The rabies virus (RABV) is a non-segmented, negative single-stranded RNA virus which causes acute infection of the central nervous system in humans. Once symptoms appear, the result is nearly always death, and to date, post-exposure prophylaxis (PEP) is the only treatment applicable only immediately after an exposure. Previous studies have identified viral RNA-dependent RNA polymerase (RdRp) as a potential drug target due to its significant role in viral replication and transcription. Herein we generated an energy-minimized homology model of RABIES-RdRp and used it for virtual screening against 2045 NCI Diversity Set III library. The best five ligand-RdRp complexes were picked for further energy minimization via molecular dynamics (MDs) where the complex with ligand Z01690699 shows a minimum score characterized with stable hydrogen bonds and hydrophobic interactions with the catalytic site residues. Our study identified an important ligand for development of remedial approach for treatment of rabies infection.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00894-021-04798-x.

Gas-solid-liquid reactive CFD simulation of an industrial RFCC riser with investigation of feed injection

The feed injection zone involving gas–solid-liquid mixing, vaporization and reaction plays an important role in fluid catalytic cracking reactors. Our previous cold-model simulation indicated that a modification from the conventionally upward feed injection scheme to a downward one improves the oil-catalyst matching. This work aims to investigate the effects of such downward feed injection on the reaction behaviors through three-phase reactive simulation of an industrial riser. A twelve-lump kinetics and vaporization of liquid oil were considered in a multiphase Eulerian model with the energy-minimization multi-scale model. The predicted solids flux, temperature, and solids concentration were compared with experimental data. The simulation showed that the downward injection with an angle of around 30° and suitable mounting position helps improving the yield of gasoline and LPG and mitigating adhesion of the coking layer to the wall. We expect such an approach sheds light on the optimization of the three-phase chemical reactors.

Strain Relaxation in "2D/2D and 2D/3D Systems": Highly Textured Mica/Bi2Te3, Sb2Te3/Bi2Te3, and Bi2Te3/GeTe Heterostructures.

Strain engineering as a method to control functional properties has seen in the last decades a surge of interest. Heterostructures comprising 2D-materials and containing van der Waals(-like) gaps were considered unsuitable for strain engineering. However, recent work on heterostructures based on Bi2Te3, Sb2Te3, and GeTe showed the potential of a different type of strain engineering due to long-range mutual straining. Still, a comprehensive understanding of the strain relaxation mechanism in these telluride heterostructures is lacking due to limitations of the earlier analyses performed. Here, we present a detailed study of strain in two-dimensional (2D/2D) and mixed dimensional (2D/3D) systems derived from mica/Bi2Te3, Sb2Te3/Bi2Te3, and Bi2Te3/GeTe heterostructures, respectively. We first clearly show the fast relaxation process in the mica/Bi2Te3 system where the strain was generally transferred and confined up to the second or third van der Waals block and then abruptly relaxed. Then we show, using three independent techniques, that the long-range exponentially decaying strain in GeTe and Sb2Te3 grown on the relaxed Bi2Te3 and Bi2Te3 on relaxed Sb2Te3 as directly observed at the growth surface is still present within these three different top layers a long time after growth. The observed behavior points at immediate strain relaxation by plastic deformation without any later relaxation and rules out an elastic (energy minimization) model as was proposed recently. Our work advances the understanding of strain tuning in textured heterostructures or superlattices governed by anisotropic bonding.

Multitarget Detection Algorithms for Multitemporal Remote Sensing Data

Target detection is always an important topic in the field of hyper/multispectral remote sensing image processing. At present, target detection algorithms in this field are generally limited to processing single-temporal remote sensing data, and they cannot obtain satisfactory results when the spectra of target and background are similar to each other. Recently, a target detection algorithm called filter tensor analysis (FTA), which is specially designed for multitemporal remote sensing data, has been reported and has achieved better detection results in many cases than the traditional single-temporal methods. However, FTA can only extract one target of interest at a time, and it cannot work when there are multiple targets of interest in the image. Therefore, considering that the matrix form of the FTA method is similar to that of the constrained energy minimization (CEM) model, it naturally comes to us that we can combine the tensor filter in FTA and the multiple target constraints to detect multiple targets by fully exploiting the time-series information in multitemporal data. To be specific, through: 1) adding the ``output to one'' constraints to the multiple targets in FTA; 2) applying linear/nonlinear function to the outputs of FTA for the multiple targets; and 3) modifying the autocorrelation matrix in FTA, four multitarget detection algorithms for multitemporal remote sensing data are proposed in this article. Experiments with simulation data and real data both show the effectiveness and superiority of the proposed methods.

A heterobimetallic tetrahedron from a linear platinum(II)-bis(acetylide) metalloligand.

Employing 4-ethynylaniline as a simple organic ligand we were able to prepare the stable trans-bis(acetylide)platinum(II) complex [Pt(L1)2(PBu3)2] as a linear metalloligand. The reaction of this metalloligand with iron(II) cations and pyridine-2-carbaldehyde according to the subcomponent self-assembly approach yielded decanuclear heterobimetallic tetrahedron [Fe4Pt6(L2)12](OTf)8. Thus, combination of these two design concepts – the subcomponent self-assembly strategy and the complex-as-a-ligand approach – ensured a fast and easy synthesis of large heterobimetallic coordination cages of tetrahedral shape with a diameter of more than 3 nm as a mixture of all three possible T-, S4- and C3-symmetric diastereomers. The new complexes were characterized by NMR and UV–vis spectroscopy and ESI mass spectrometry. Using GFN2-xTB we generated energy-minimized models of the diastereomers of this cage that further corroborated the results from analytical findings.

Fundamental Insight into the Dripping Criteria of Slag and Hot Metal in a Coke Bed Using Energy Minimization Model

The dripping zone of the blast furnace is important in determining the productivity and quality of hot metal. The permeability of the bed affects both of these factors. Static holdup of hot metal and slag resulting from interfacial phenomena adversely affect the permeability of the bed. To characterize the holdup in the packed bed, we assumed the packed bed as an ensemble of two-sphere contacts at various contact axis inclinations and for contact angles. We developed mathematical models to determine the shapes of stable films at these contacts taking into account interfacial and gravitational energies. The film at higher contact angles shows an asymmetric shape even in between vertical spheres, which is investigated in detail. We also show the reason for the difficulty for dripping of the liquids at smaller particle sizes.

A Novel Speckle Noise Removal Algorithm Based on ADMM and Energy Minimization Method

Speckle noise removal in medical ultrasound images is a challenging task. In this paper, a new model is proposed to removal speckle noise, alternating direction method of multipliers algorithm is employed to solve the new energy minimization model. The convexity, existence, and uniqueness of the new energy minimization model’s solution are proved. Series of experiments are designed in this paper. Numerical results show that the new algorithm can reduce the step effect effectively obtain good results in visual effect and quantitative measures by comparing with some traditional models.