Adaptive Partitioning Research Articles

Influence of Magnetic Polarization on Diagrams of Inductive and Electromagnetic Logs

Abstract ––We present the results of the study of the effect of induced magnetic polarization of clay beds under the influence of an external harmonic electromagnetic field (frequencies 70 and 875 kHz). A two-stage numerical modeling procedure is proposed. At the first stage we determine the effective relative magnetic permeability of a synthetic sample with inclusions of clay particles. In this case a 3D heterogeneous mesh sample is generated. Then we numerically model a spatial distribution of an electric field. The electromotive force (EMF) induced in the measuring coil is calculated from this distribution. Relative magnetic permeability is determined by comparison with EMF of homogeneous samples with different values of magnetic permeability. It has been found that during the electric field excitation by an alternating current coil, the effect of induced magnetic polarization appears in the sample with clay particles. Its manifestation is that the effective magnetic permeability becomes complex. At the second stage we calculate the EMF diagram of the three-coil logging probe in the macro-model ‘clay cap – reservoir’. Magnetic permeability of the clay cap is given by a complex value. In the generated logs, extremes appear in the vicinity of the bottom of the clay cap; they do not correspond to the distribution of electrical conductivity and magnetic permeability in the given model. They can be incorrectly interpreted when analyzing real logs into individual formations. Numerical modeling at all stages is performed by the Vector Finite Element Method on a consistent adaptive tetrahedral partition and the first-order Webb vector basis.

A multi-scale adaptive grid partition method based on two-dimensional Fourier transform for ZTD

A multi-scale adaptive grid partition method based on two-dimensional Fourier transform for ZTD

Nonplanar Aromaticity of Dinuclear Rare-Earth Metallacycles.

While the concept of metalla-aromaticity has well been extended to transition organometallic compounds in diverse geometries, aromatic rare-earth organometallic complexes are rare due to the special (n - 1)d0 configuration and high-lying (n - 1)d orbitals of rare-earth centers. In particular, nonplanar cases of rare-earth complexes have not been reported so far. Here, we disclose the nonplanar aromaticity of dinuclear scandium and samarium metallacycles characterized by various aromaticity indices (nucleus-independent chemical shift, isochemical shielding surface, anisotropy of induced current density, and isomerization stabilization energy). Bonding analyses (Kohn-Sham molecular orbital, adaptive natural density partitioning, multicenter bond indices, and principal interacting orbital) reveal that three delocalized π orbitals, predominantly contributed by the 2-butene tetraanion ligand, result in the formation of six-electron conjugated systems. Guided by these findings, we predicted that the lutetium and gadolinium analogues of dinuclear rare-earth metallacycles should be aromatic, which have been verified by the successful synthesis of real molecules. This work extends the concept of nonplanar aromaticity to the field of rare-earth metallacycles and illuminates the path for designing and synthesizing various rare-earth metalla-aromatics.

Bi-directional ensemble differential evolution for global optimization

Hybridizing multiple mutation strategies has shown much effectiveness in helping differential evolution (DE) algorithms achieve good optimization performance. Though abundant adaptive ensemble strategies have been developed to adaptively employ multiple mutation strategies to evolve the population, most of them ignore to make full use of the properties and characteristics of the multiple mutation strategies. To fill this gap, this paper devises a bi-directional ensemble scheme for DE to adaptively assemble totally 8 mutation strategies with different properties and characteristics. As a result, a novel DE, which we call bi-directional ensemble DE (BDEDE), is developed. Specifically, this paper sorts the 8 mutation strategies roughly from two opposite perspectives, namely the convergence and the diversity. Then, we assign each mutation strategy with two different non-linear probabilities, which are calculated on the basis of its two rankings obtained from the two perspectives. Subsequently, to make full use of these mutation schemes, we first partition the whole population into two separate parts, namely elite individuals and non-elite individuals. Then, for each elite individual, we randomly select a mutation strategy from the 8 candidates based on the probabilities calculated by the convergence rankings, while for each non-elite individual, we stochastically choose a mutation scheme from the same 8 candidates but based on the probabilities computed by the diversity rankings. In this manner, the elite individuals prefer to exploit the located optimal areas, while the non-elite individuals tend to explore the solution space. Therefore, it is likely that BDEDE expectedly maintains a good balance between search diversity and search convergence. To further help BDEDE achieve such a purpose, this paper devises an adaptive partition strategy to dynamically separate the whole population into the two categories. With the above two techniques, BDEDE anticipatedly obtains good optimization performance. To verify its effectiveness and efficiency, we conduct experiments on the CEC2014 and the CEC2017 benchmark sets by comparing BDEDE with totally 14 well-known and state-of-the-art DE variants. Experimental results have shown that BDEDE performs competitively with or even significantly better than the 14 compared DE variants. The source code of BDEDE can be downloaded from https://gitee.com/mmmyq/BDEDE.

Restriction on molecular fluxionality by substitution: A case study for the 1,10-dicyanobullvalene.

We show herein that 1,10-dicyano substitution restricts the paragon fluxionality of bullvalene to just 14 isomers which isomerize along a single cycle. The restricted fluxionality of 1,10-dicyanobullvalene (DCB) is investigated by means of: (i) Bonding analyses of the isomer structures using the adaptive natural density partitioning (AdNDP). (ii) Quantum dynamical simulations of the isomerizations along the cyclic intrinsic reaction coordinate of the potential energy surface (PES). The PES possesses 14 equivalent potential wells supporting 14 isomers which are separated by 14 equivalent potential barriers supporting 14 transition states. Accordingly, at low temperatures, DCB appears as a hindered molecular rotor, without any delocalization of the wavefunction in the 14 potential wells, without any nuclear spin isomers, and with completely negligible tunneling. These results are compared and found to differ from those for molecular boron rotors. (iii) Born-Oppenheimer molecular dynamics (BOMD) simulations of thermally activated isomerizations. (iv) Calculations of the rate constants in the frame of transition state theory (TST) with reasonable agreement achieved with the BOMD results. (v) Simulations of the equilibration dynamics using rate equations for the isomerizations with TST rate coefficients. Accordingly, in the long-time limit, isomerizations of the 14 isomers, each with Cs symmetry, approach the "14 Cs → C7v" thermally averaged structure. This is a superposition of the 14 equally populated isomer structures with an overall C7v symmetry. By extrapolation, the results for DCB yield working hypotheses for so far un-explored properties e.g. for the equilibration dynamics of C10H10.

Effective anti-submarine decision support system based on heuristic rank-based Dijkstra and adaptive threshold partitioning mechanism

Submarines possess strong covert striking capabilities, making anti-submarine warfare (ASW) a global naval priority. The Hidden Markov Anti-Submarine Model (HMASM) finds crucial applications in dynamic and uncertain ASW. The model delineates ASW into two phases: partitioning and search path planning. Current partitioning algorithms often include cells leading to redundant and competitive search, determined empirically. Additionally, in HMASM, the path planning algorithms based on genetic algorithm (GA) are time-consuming and exhibits unstable performance. This paper proposes a novel solution to these issues. For partitioning, a Reassigned k-Nearest Neighbour (RKNN) algorithm is introduced, identifying and reallocating units causing repeated and competing searches. For search path planning, heuristic rules for Dijkstra's cost function and goal point selection transform the NP-hard problem of maximizing the expected number of detections (ED) into a deterministic algorithm-compatible form. A heuristic rank-based selection model, Rank-Based Dijkstra under the Hidden Markov Model (HMM-R-Dijkstra), considering distance and probability, is added to Dijkstra. Furthermore, an Adaptive Threshold Partitioning (ATP) dynamically monitors searcher exploration by setting variables and thresholds to determine optimal partitioning timing, preventing untimely and excessive partitioning. Combining RKNN, HMM-R-Dijkstra, and ATP forms R-HRD-ATP, optimizing all parameters using parallel structures. Through three comparative experiments, R-HRD-ATP's performance steadily improves. Experiments comparing R-HRD-ATP to GA, Sparrow Search Algorithm (SSA)-GA, and Ant Colony Optimization (ACO)-GA reveal performance enhancements of 25–70.99% and time savings of 56–295 times for our model. Importantly, no parameter adjustments are required. The success of R-HRD-ATP indicates that its heuristic rules can support the establishment of robust applications of deterministic path planning algorithms in HMASM.

A Precision Benchmark Suite for Nuclear Reactor Point Kinetics Equations via Converged Accelerated Taylor Series (CATS)

Extreme benchmarks of 10 or more places for the point kinetics equations for time-dependent nuclear reactor power transients are rare. Therefore, to establish an extreme benchmark, we employ a Taylor series (TS) with continuous analytical continuation to solve the ordinary differential equations of point kinetics including feedback. Nonlinear Wynn-epsilon convergence acceleration confirms the highly precise solutions for neutron and precursor densities. Through adaptive partitioning of time intervals, the proposed Converged Accelerated Taylor Series, or CATS algorithm in double precision, automatically performs successive mesh refinement to obtain high-precision initial conditions for each subinterval, with the intent to reduce propagation error. Confirmation of 10 to 12 places comes from comparison to the BEFD (Backward Euler Finite Difference) algorithm in quadruple precision also developed by the author. We report benchmark results for common cases found in the literature including step, ramp, zigzag, and sinusoidal prescribed reactivity insertions and insertions with nonlinear adiabatic Doppler feedback. We also establish a suite of new prescribed reactivity insertions and insertions with feedback, based on reactivities with Taylor series representations as suggested by the CATS algorithm.

A multiscale fracture model using peridynamic enrichment of finite elements within an adaptive partition of unity: Experimental validation

Partition of unity methods (PUM) are of domain decomposition type and provide the opportunity for multiscale and multiphysics numerical modeling. Here, we apply Peridynamic (PD) enrichment to propagate cracks in the PUM global–local enrichment scheme. We apply linear elasticity globally and PD over local zones where fractures occur. The elastic fields provide appropriate boundary data for local PD simulations on a subdomain containing the crack tip to grow the crack. Once the updated crack path is found the elastic field in the body and surrounding the crack is updated using the PUM basis with an elastic field enrichment near the crack. The subdomain for the PD simulation is chosen to include the current crack tip as well as features that influence crack growth. This paper is part II of this series and validates the combined PD/PUM simulator against the experimental results. The results of numerical simulation show that we attain good agreement between experiment and simulation with a local PD subdomain that is moving with the crack tip and adaptively chosen size.

Introducing hafnium to atomically small- and medium-sized tin clusters (HfSnn0/-/2- (n = 4–17)): A computational investigation of geometrical and growth behavior, spectral properties, electronic configuration and thermochemistry

The global and local minimum configurations of single Hf atom doped Sn clusters are conducted via density function theory (DFT) combined with artificial bee colony algorithm (ABCluster). Furthermore, DFT method is also used to systematically investigate on their structural growth evolution, spectral and electronic information, thermochemical properties following the size of tin clusters doped Hf atom. Structurally, the ground-state geometries of neutral, anion and di-anion are discovered that, from n = 4, the number of Sn atoms in cluster, HfSnn0/-/2- adsorb additional Sn atom on the prior architecture one by one until forming n = 17 for HfSnn-10/-, as well as forming n = 16 for HfSnn-12-. And for the HfSn110/- and HfSn102- as beginning the species veritably develop sealed architectures. The strongest vibrational modes of sealed nanoclusters are stretching modes of Hf atom with infrared actives and breathing modes of the Sn cage framework with Raman actives, respectively. The natural population analysis (NPA) elucidates the stronger relationship between the Hf atoms and the tin frameworks in sealed clusters than that in unsealed clusters. The results of thermochemical properties, molecular orbital shell (MOs), adaptive natural density partitioning (AdNDP) and ultraviolet visible absorption spectrum (UV–Vis) indicate that, the HfSn16 with high symmetry of Td exhibits thermochemical stability and optoelectronic properties, which is utilized potentially as zero-dimensional unit of self-assembling fluorescent nanomaterials.

Game-Based Adaptive FLOPs and Partition Point Decision Mechanism With Latency and Energy-Efficient Tradeoff for Edge Intelligence

Game-Based Adaptive FLOPs and Partition Point Decision Mechanism With Latency and Energy-Efficient Tradeoff for Edge Intelligence

Learning Only When It Matters: Cost-Aware Long-Tailed Classification

Most current long-tailed classification approaches assume the cost-agnostic scenario, where the training distribution of classes is long-tailed while the testing distribution of classes is balanced. Meanwhile, the misclassification costs of all instances are the same. On the other hand, in many real-world applications, it is more proper to assume that the training and testing distributions of classes are the same, while the misclassification cost of tail-class instances is varied. In this work, we model such a scenario as cost-aware long-tailed classification, in which the identification of high-cost tail instances and focusing learning on them thereafter is essential. In consequence, we propose the learning strategy of augmenting new instances based on adaptive region partition in the feature space. We conduct theoretical analysis to show that under the assumption that the feature-space distance and the misclassification cost are correlated, the identification of high-cost tail instances can be realized by building region partitions with a low variance of risk within each region. The resulting AugARP approach could significantly outperform baseline approaches on both benchmark datasets and real-world product sales datasets.

AMSPM: Adaptive Model Selection and Partition Mechanism for Edge Intelligence-driven 5G Smart City with Dynamic Computing Resources

With the help of 5G network, edge intelligence (EI) can not only provide distributed, low-latency, and high-reliable intelligent services, but also enable intelligent maintenance and management of smart city. However, the constantly changing available computing resources of end devices and edge servers cannot continuously guarantee the performance of intelligent inference. In order to guarantee the sustainability of intelligent services in smart city, we propose the Adaptive Model Selection and Partition Mechanism (AMSPM) in 5G smart city where EI provides services, which mainly consists of Adaptive Model Selection (AMS) and Adaptive Model Partition (AMP). In AMSPM, the model selection and partition of deep neural network (DNN) are formulated as an optimization problem. Firstly, we propose a recursive-based algorithm named AMS based on the computing resources of edge devices to derive an appropriate DNN model that satisfies the latency demand of intelligent services. Then, we adaptively partition the selected DNN model according to the computing resources of edge devices. The experimental results demonstrate that, when compared with state-of-the-art model selection and partition mechanisms, AMSPM not only reduces latency but also enhances computing resource utilization.

Unveiling the structural and bonding properties of AuSi2- and AuSi3- clusters: A comprehensive analysis of anion photoelectron spectroscopy and abinitio calculations.

Silicon clusters infused with transition metals, notably gold, exhibit distinct characteristics crucial for advancing microelectronics, catalysts, and energy storage technologies. This investigation delves into the structural and bonding attributes of gold-infused silicon clusters, specifically AuSi2- and AuSi3-. Utilizing anion photoelectron spectroscopy and abinitio computations, we explored the most stable isomers of these clusters. The analysis incorporated Natural Population Analysis, electron localization function, molecular orbital diagrams, adaptive natural density partitioning, and Wiberg bond index for a comprehensive bond assessment. Our discoveries reveal that cyclic configurations with the Au atom atop the Si-Si linkage within the fundamental Si2 and Si3 clusters offer the most energetically favorable structures for AuSi2- and AuSi3- anions, alongside their neutral counterparts. These anions exhibit notable highest occupied molecular orbital-lowest unoccupied molecular orbital gaps and significant σ and π bonding patterns, contributing to their chemical stability. Furthermore, AuSi2- demonstrates π aromaticity, while AuSi3- showcases a distinctive blend of σ antiaromaticity and π aromaticity, crucial for their structural robustness. These revelations expand our comprehension of gold-infused silicon clusters, laying a theoretical groundwork for their potential applications in high-performance solar cells and advanced functional materials.

Model-agnostic progressive saliency map generation for object detector

With the widespread adoption of object detection models across various industries, the interpretability of these detectors has become an important research topic. The interpretability of a detector helps humans understand which areas significantly contribute to the model's decision. Furthermore, the interpretability enhances the credibility of detectors and helps identify their strengths and weaknesses. Due to the ability to provide intuitive explanations of models, the saliency map has been widely employed in the field of interpreting deep models. Model-agnostic interpretability methods are more general approaches as they treat the model as a black box without considering its internal complexity structure. However, existing model-agnostic interpretability methods often introduce “noise” into saliency maps by applying random masking and fixed masking granularity. This noise reduces the quality and interpretability of the generated saliency maps. To address this challenge and obtain more interpretable saliency maps for object detection models, this paper proposes a model-agnostic progressive saliency map generation method based on a hierarchical framework called MAPSM. In MAPSM, an adaptive masking partition mechanism is introduced to adapt the masking granularity to different object sizes. Additionally, MAPSM employs a saliency-driven mask generation strategy to effectively reduce the “noise”. Utilizing a hierarchical framework, MAPSM progressively discovers and refines the saliency areas of objects, resulting in more interpretable saliency maps. To evaluate the quality of the saliency maps generated by MAPSM, we compare it with other methods in multiple metrics. Experimental results demonstrate that our method produces saliency maps with better quality and interpretability.

Tree Learning: Towards Promoting Coordination in Scalable Multi-Client Training Acceleration

Iteration based collaborative learning (CL) paradigms, such as federated learning (FL) and split learning (SL), faces challenges in training neural models over the rapidly growing yet resource-constrained edge devices. Such devices have difficulty in accommodating a full-size large model for FL or affording an excessive waiting time for the mandatory synchronization step in SL. To deal with such challenge, we propose a novel CL framework which adopts an tree-aggregation structure with an adaptive partition and ensemble strategy to achieve optimal synchronization and fast convergence at scale. To find the optimal split point for heterogeneous clients, we also design a novel partitioning algorithm by minimizing the idleness during communication and achieving the optimal synchronization between clients. In addition, a parallelism paradigm is proposed to unleash the potential of optimum synchronization between the clients and server to boost the distributed training process without losing model accuracy for edge devices. Furthermore, we theoretically prove that our framework can achieve better convergence rate than state-of-the-art CL paradigms. We conduct extensive experiments and show that our framework is 4.6× in training speed as compared with the traditional methods, without compromising training accuracy.

Semantic-aware automatic extraction method for bottom sediment annotations in raster nautical charts

ABSTRACT The automatic extraction of bottom sediment annotations in large-scale raster nautical charts has limitations, including an imprecise semantic information description and low efficiency. To overcome them, we propose a convolutional neural network (CNN)-based method for the automatic extraction of bottom sediment annotations in raster nautical charts, using image processing techniques to improve it. First, an adaptive chart partitioning model that considers element completeness is constructed. Second, a principle for the unique identification of elements based on spatial conflicts is designed. Finally, a model for accurately extracting semantic information for bottom sediment annotations is established. To evaluate the effectiveness of the proposed method, we implemented a model based on the PyTorch framework and used the PIL library to analyze the results. We also conducted comparative experiments on multiple CNN models to recommend the selection of such models in the proposed method by comparing their classification and recognition performance. The experimental results indicate that (1) the proposed model can achieve high-precision extraction of bottom sediment annotations in raster nautical charts. (2) Furthermore, the proposed model generally has high recognition accuracy and semantic completeness, with better recognition precision than traditional pattern recognition methods.

The geometry, electronic and magnetic properties of Zr n Ni (n = 2-14) clusters: A first-principles jointed particle-swarm-optimization investigation

Zirconium-nickel binary alloys and metal glass have superior performance like ultrahigh fracture strength, good toughness. In this paper, the structures of small-sized Zr n Ni (n = 2–14) clusters have been searched using the particle swarm algorithm in combination with density-functional theory (DFT). The geometrical configuration tends to form a three-dimensional structure as the number of atoms in the cluster increases. By calculating the average binding energy per atom, second-order difference of energy, and dissociation energy of Zr n Ni (n = 2–14) clusters, it is demonstrated that Zr n Ni (n = 7, 12) clusters are more stable than their neighbors, and can be used as a candidate structure for magic number clusters. The electron localization function (ELF) calculations reveal those metallic bonds of Zr-Ni and Zr-Zr atoms. The Adaptive natural density partitioning results show that there are 20 three-center and 7 seven-center two-electron orbitals which make the quenching of magnetic moments of Zr12Ni atoms.

FeC4H22+ Encompassing Planar Tetracoordinate Iron: Structure and Bonding Patterns

The singlet, triplet, and quintet electronic states of the FeC4H22+ system are theoretically explored using quantum chemical methods, and 39 isomers are identified in the singlet electronic state and 4 isomers in both triplet and quintet electronic states. A molecule with a planar tetracoordinate iron (ptFe) is found on the potential energy surface of singlet and triplet electronic states. The bonding features of ptFe in the singlet electronic state are analyzed with natural bond orbital (NBO) analysis, adaptive natural density partitioning (AdNDP), and molecular orbital analysis. The resultant data delineate that the ptFe is stabilized through electron delocalization in the ptFe system.

Data-based equations for damage evaluation of reinforced concrete columns with square cross sections

This study aims to develop simple equations for assessing the damage of reinforced concrete (RC) columns subjected to seismic loading. A fiber element model in OpenSees, validated against experimental data, was used to generate a large dataset of 56548 data values for various RC columns. The Monte Carlo Hierarchical Adaptive Random Partitioning (MC-HARP) strategy was then applied to evaluate the sufficiency of the dataset and propose reliable equations for estimating the residual displacement ratio (RDR), damage index (DI), and residual curvature of RC columns under given displacement demands. A relationship between these parameters and the structural performance levels was also established. Residual displacement is a practical measure that can be observed in the field after an earthquake. The proposed equations facilitate the determination of the performance level of the structure after an earthquake event. The findings of this paper can also be used in designing new structures within a performance-based seismic design (PBSD) context.

The structural and electronic split: Boron vs aluminum hydrides

We systematically investigated the structural evolution of boron (B) and aluminum (Al) hydrides using various DFT and ab initio methods, aiming to reveal the similarities and differences in their geometric and electronic structures. While B hydrides have been extensively studied both experimentally and theoretically, less is known about its group 13 heavier congener, Al. Extensive global minimum searches of the B2Hx (Al2Hx) and B3Hy (Al3Hy) hydrides (x = [0–6], y = [0–9]) were performed to identify the most stable geometric structures for each stoichiometry. In most of the series, B and Al hydrides exhibit qualitatively different structures, except for the most saturated X2H5 and X2H6 stoichiometries. Chemical bonding analyses employing adaptive natural density partitioning and electron localization function methods identified notable differences between B and Al hydrides in most of the compositions. B hydrides predominantly possess two-center (2c) and three-center (3c) bonding elements, suggesting a relatively balanced electron distribution. On the contrary, Al hydrides tend to retain unpaired electrons or lone pairs on Al atoms, forming a large number of closely lying isomers with various combinations of 1c, 2c, 3c, and 4c bonding elements. Thermodynamic stability analyses revealed that all studied clusters demonstrated stability toward various H/H2 dissociation pathways, with Al hydrides being less stable than B counterparts.