Large-scale Simulations Research Articles

Balancing simulation performance and computational intensity of CA models for large-scale land-use change simulations

Balancing simulation performance and computational intensity of CA models for large-scale land-use change simulations

Gradient projection method for enforcing crack irreversibility as box constraints in a robust monolithic phase-field scheme

A phase-field monolithic scheme based on the gradient projection method is developed to model crack propagation in brittle materials under cyclic loading. As a type of active set method, the gradient projection method is particularly attractive to enforce the irreversibility condition imposed on the phase-field variables as bound constraints, or box constraints. This method has the advantages of allowing the rapid change of active constraints during iterations and computing the projected gradient with a negligible cost. The gradient projection method is further combined with the limited-memory BFGS (L-BFGS) method to overcome the convergence difficulties arising from the non-convex energy functional. A compact representation of the BFGS matrix is adopted as the limited-memory feature to avoid the storage of fully dense matrices, making this method practical for large-scale finite element simulations. By locating the generalized Cauchy point on the piecewise linear path formed by the projected gradient, the active set of box constraints can be determined. The variables in the active set, which are at the boundary of the box constraints, are kept fixed to form a subspace minimization problem. A primal approach and a dual approach are presented to solve this subspace minimization problem for the remaining free variables at the generalized Cauchy point. Several two-dimensional (2D) and three-dimensional (3D) examples are provided to demonstrate the capabilities of the proposed monolithic scheme, particularly in enforcing the phase-field irreversibility during crack propagation under cyclic loading. In these numerical examples, the proposed monolithic scheme is combined with an adaptive mesh refinement technique to alleviate the heavy computational cost incurred by the fine mesh resolution required around the crack region. The proposed method is further compared with two other phase-field solving techniques regarding the convergence behavior. To ensure a fair comparison, the same problem settings and implementation techniques are adopted. The proposed monolithic scheme provides a unified framework to overcome the numerical difficulties associated with the non-convex energy functional, effectively enforce the phase-field irreversibility to ensure the thermodynamic consistency, and alleviate the heavy computational cost through adaptive mesh refinement in 2D and 3D phase-field crack simulations.

Hamiltonian learning for 300 trapped ion qubits with long-range couplings.

Quantum simulators with hundreds of qubits and engineerable Hamiltonians have the potential to explore quantum many-body models that are intractable for classical computers. However, learning the simulated Hamiltonian, a prerequisite for any quantitative applications of a quantum simulator, remains an outstanding challenge due to the fast increasing time cost with the qubit number and the lack of high-fidelity universal gate operations in the noisy intermediate-scale quantum era. Here, we demonstrate the Hamiltonian learning of a two-dimensional ion trap quantum simulator with 300 qubits. We use global manipulations and single-qubit-resolved state detection to efficiently learn the all-to-all-coupled Ising model Hamiltonian, with the required quantum resources scaling at most linearly with the qubit number. We further demonstrate a physically guided learning scheme with the quantum sample complexity independent of system sizes by carefully fitting the anharmonic trap potential. Our work paves the way for wide applications of large-scale ion trap quantum simulators.

The role of dry mergers in shaping the scaling relations of galaxies

In the context of the hierarchical formation of galaxies, we investigated the role played by mergers in shaping the scaling relations of galaxies, that is the projections of their fundamental plane onto the and planes. To this end, based on the scalar virial theorem, we developed a simple theory of multiple dry mergers to read both the large-scale simulations and the companion scaling relations. The aim was to compare the results of this approach with the observational data and with two of the most recent and detailed numerical cosmo-hydro-dynamical simulations: Illustris-TNG and EAGLE (Evolution and Assembly of GaLaxies and their Environments). We derived these scaling relations for the galaxies of the Mapping Nearby Galaxies at APO (MaNGA) and Wide-field Imaging of Nearby Galaxy-Clusters Survey (WINGS) databases and compared them with the observational data, the numerical simulations, and the results of our simple theory of dry mergers. The multiple dry merging mechanism is able to explain all the main characteristics of the observed scaling relations of galaxies, such as slopes, scatters, curvatures, and zones of exclusion. The distribution of galaxies in these planes is continuously changing across time because of the merging activity and other physical processes, such as star formation, quenching, and energy feedback. The simple merger theory presented here yields the correct distribution of galaxies in the main scaling relations at all cosmic epochs. The precision is comparable with that obtained by the modern cosmo-hydro-dynamical simulations, with the advantage of providing a rapid exploratory response on the consequences engendered by different physical effects.

Numerical optimization of large-scale monotone equations using the free-derivative spectral conjugate gradient method

This study introduced an efficient method for solving non-linear equations. Our approach enhances the traditional spectral conjugate gradient parameter, resulting in significant improvements in the resolution of complex nonlinear problems. This innovative technique ensures global convergence and descent condition supported by carefully considered assumptions. The efficiency and effectiveness of the proposed method is highlighted by its outstanding numerical performance. To validate our claims, large-scale numerical simulations were conducted. These tests were designed to evaluate the capabilities of our proposed algorithm rigorously. In addition, we performed a comprehensive comparative numerical analysis, benchmarking our method against existing techniques. This analysis revealed that our approach consistently outperformed others in terms of theoretical robustness and numerical efficiency. The superiority of our method is evident in its ability to solve large-scale problems with accuracy in function evaluations, fewer iterations, and improved computational performance thereby, making it a valuable contribution to the field of numerical optimization.

Is the James Webb Space Telescope detecting too many AGN candidates?

Abstract In less than two years of operation, the James Webb Space Telescope (JWST) has already accelerated significantly our quest to identify active massive black holes (BHs) in the first billion years of the Universe’s history. At the time of writing, about 50 AGN detections and candidates have been identified through spectroscopy, photometry, and/or morphology. Broad-line AGN are about a hundred times more numerous than the faint end of the UV-bright quasar population at z ∼ 5 − 6. In this paper, we compare the observational constraints on the abundance of these AGN at z ∼ 5 to the populations of AGN produced in large-scale cosmological simulations. Assuming a null fraction of obscured simulated AGN, we find that while some simulations produce more AGN than discovered so far, some others produce a similar abundance or even fewer AGN in the bolometric luminosity range probed by JWST. Keeping in mind the large uncertainty on the constraints, we discuss the implications for the theoretical modeling of BH formation and evolution in case similar constraints continue to accumulate. At the redshift of interest, the simulated AGN populations diverge the most at Lbol ∼ 1044 erg s−1 (by more than a dex in the bolometric luminosity function). This regime is most affected by incompleteness in JWST surveys. However, it holds significant potential for constraining the physical processes determining the assembly of BHs (e.g. seeding, feedback from supernova and AGN) and the current abundance of broad-line AGN with ≥1044.5 erg s−1.

Real-Time Power Management of Plug-in Electric Vehicles and Renewable Energy Sources in Virtual Prosumer Networks with Integrated Physical and Network Security Using Blockchain

This paper presents a blockchain-enabled Multi-Agent System (MAS) for real-time power management in Virtual Prosumer (VP) Networks, integrating Plug-in Electric Vehicles (PEVs) and Renewable Energy Sources (RESs). The proposed framework addresses critical challenges related to scalability, security, and operational efficiency by developing a hierarchical MAS architecture that optimizes the scheduling and coordination of geographically distributed PEVs and RESs. Unlike conventional business management systems, this study integrates a blockchain-based security mechanism within the MAS framework, leveraging Proof of Authority (PoA) consensus to enhance transaction security while addressing scalability and energy consumption concerns. The system further employs advanced Particle Swarm Optimization (PSO) to dynamically compute optimal power set-points, enabling adaptive and efficient energy distribution. Additionally, hierarchical aggregation of transactions at lower MAS layers enhances computational efficiency and system resilience under high-traffic and partial network failure conditions. The proposed framework is validated through large-scale simulations spanning four major cities in Greece, demonstrating its scalability, reliability, and efficiency under diverse operational scenarios. Results confirm that the system effectively balances energy supply and demand while maintaining secure and transparent transactions. Despite these advancements, practical deployment challenges remain, including synchronization delays in geographically distributed agents, legacy system integration, and blockchain energy consumption. Future research directions include investigating more advanced consensus mechanisms (e.g., Proof of Task), machine learning-driven predictive optimization, real-world large-scale testing, and federated learning models for decentralized decision-making. The proposed framework offers a scalable, secure, and efficient solution for decentralized real-time energy management in Virtual Prosumer Networks.

A Monte Carlo algorithm for large-scale magnetic simulations

PurposeThe paper refers to Monte Carlo (MC) magnetic simulations and finite element scaling rules, allowing the simulation of magnetic mesoscopic systems. The scaling rules consist of a redefinition of the system Hamiltonian as a function of a scaling factor, including spin values and exchange integral parameters attributed to system nodes. This approach leads to a disturbance of the thermodynamic balance expressed by acceptance probability in the Metropolis steps. This paper aims to explain the effect and solve the problem.Design/methodology/approachThe change of the acceptance probability is derived from energy change and the scaling roles. As a test, a set of thermomagnetic curves was simulated in the temperature ranges related to ferro-paramagnetic transitions for the systems with different enlargement factors.FindingsThe presented approach allows the simulation of thermomagnetic curves of large magnetic systems using the MC Metropolis algorithm modified by the scaling rules.Originality/valueThe presented approach allows the simulation of thermomagnetic curves of large magnetic systems, using the MC Metropolis algorithm modified by the scaling rules.

Improved streamflow simulations in hydrologically diverse basins using physically informed deep learning models

Physically informed deep learning models, especially Long Short-Term Memory (LSTM) networks, have shown promise in large-scale streamflow simulations. However, an in-depth understanding of the relative contribution of physical information in deep learning models has been missing. Using a large-sample testbed of 220 catchments in hydrologically diverse regions of the Indian subcontinent, we quantify the impact of incremental addition of physical information on model performance using multiple variants of the LSTM model based on various combinations of static catchment attributes and simulated land surface states. We found that LSTM models trained with catchment geophysics as additional input outperformed the base LSTM model in terms of the nationwide median Kling-Gupta Efficiency (KGE) of in-sample catchments, increasing KGE from 0.32 to 0.60. Additionally, the model retained significant prediction skill in out-of-sample catchments, demonstrating that a pre-trained LSTM model can be a powerful tool to predict streamflow in data-scarce regions.

Cross-Point Ferroelectric Hf0.5Zr0.5O2 Capacitors for Remanent Polarization-Driven In-Memory Computing.

Ferroelectric Hf0.5Zr0.5O2 (HZO) capacitors have been extensively explored for in-memory computing (IMC) applications due to their nonvolatility and back-end-of-line (BEOL) compatible process. Several IMC approaches using resistance and capacitance states in ferroelectric HZO have been proposed for vector-matrix multiplication (VMM), but previous approaches suffer from limited accuracy and reliability. In this work, we propose a promising approach centered on the remanent polarization (Pr) switching of binary ferroelectric HZO capacitor synapses. We experimentally demonstrate a simple pattern recognition task showing that the voltage readout of Pr switching provides excellent accuracy due to its high on/off ratio and consequent reliability. We also performed large-scale simulations on a complex inference task, achieving high accuracy and immunity to device variations. We therefore believe that our proposed paradigm is promising for near-term neuromorphic IMC.

Prediction can be safely used as a proxy for explanation in causally consistent Bayesian generalized linear models

Bayesian modeling provides a principled approach to quantifying uncertainty and has seen a surge of applications in recent years. Within the context of a Bayesian workflow, we are concerned with model selection for the purpose of finding models that best explain the data or underlying data generating process. Since insight into the true process is rare, what remains is incomplete causal knowledge and model predictions of the data. This leads to the important question of when the use of prediction as a proxy for explanation for the purpose of model selection is valid. We approach this question by means of large-scale simulations of Bayesian generalized linear models where we investigate various causal and statistical misspecifications. Our results indicate that the use of prediction as proxy for explanation is valid and safe if the models under consideration are sufficiently consistent with the underlying causal structure of the true data generating process.

Anomalous entropy-driven kinetics of dislocation nucleation

The kinetics of dislocation reactions, such as dislocation multiplication, controls the plastic deformation in crystals beyond their elastic limit, therefore critical mechanisms in a number of applications in materials science. We present a series of large-scale molecular dynamics simulations that shows that one such type of reactions, the nucleation of dislocation at free surfaces, exhibit unconventional kinetics, including unexpectedly large nucleation rates under compression, very strong entropic stabilization under tension, as well as strong non-Arrhenius behavior. These unusual kinetics are quantitatively rationalized using a variational transition state theory approach coupled with an efficient numerical scheme for the estimation of vibrational entropy changes. These results highlight the need for a variational treatment of the kinetics to quantitatively capture dislocation reaction kinetics, especially at low-to-moderate strains where large deformations are required to activate reactions. These observations suggest possible explanations to previously observed unconventional deformation kinetics in both molecular dynamics simulations and experiments.

Gapless topological behaviors in a long-range quantum spin chain

Topology is a fascinating phenomena in condensed-matter physics typically associated with a bulk gap. However, recent research shifts focus to quantum critical points or phases that exhibit nontrivial topological properties. Here we explore a cluster-Ising chain with long-range antiferromagnetic interactions that decay as a power law with distance. Using large-scale density matrix renormalization group simulations, we demonstrate that the nontrivial topology at the critical point remains stable against long-range interactions, resulting in a topologically nontrivial critical line. Moreover, even within the gapped region, the interplay between topology and long-range interaction can give rise to a topological phase featuring algebraically decaying correlations and edge modes, similar to gapless topological phases. We refer to this phase as the algebraic topological phase, which exhibits nontrivial gapless topological behaviors and arises solely from long-range interactions without short-range counterparts. The findings pave the way for more studies on topological states in long-range systems.

Systematic Approach to Parametrization of Disaccharides for the Martini 3 Coarse-Grained Force Field.

Sugars are ubiquitous in biology; they occur in all kingdoms of life. Despite their prevalence, they have often been somewhat neglected in studies of structure-dynamics-function relationships of macromolecules to which they are attached, with the exception of nucleic acids. This is largely due to the inherent difficulties of not only studying the conformational dynamics of sugars using experimental methods but indeed also resolving their static structures. Molecular dynamics (MD) simulations offer a route to the prediction of conformational ensembles and the time-dependent behavior of sugars and glycosylated macromolecules. However, at the all-atom level of detail, MD simulations are often too computationally demanding to allow a systematic investigation of molecular interactions in systems of interest. To overcome this, large scale simulations of complex biological systems have profited from advances in coarse-grained (CG) simulations. Perhaps the most widely used CG force field for biomolecular simulations is Martini. Here, we present a parameter set for glucose- and mannose-based disaccharides for Martini 3. The generation of the CG parameters from atomistic trajectories is automated as fully as possible, and where not possible, we provide details of the protocol used for manual intervention.

An Integral-Direct GOSTSHYP Algorithm for the Computation of High Pressure Effects on Molecular and Electronic Structure.

To simulate the effects of high pressure on molecular and electronic structure, methods based on the polarizable continuum model have emerged as a serious contender to the conventionally employed periodic boundary conditions. In this work, we present a highly efficient integral-direct algorithm for the Gaussians On Surface Tesserae Simulate HYdrostatic Pressure (GOSTSHYP) method. We examine the efficiency of this implementation on large chains of α-d-glucose units. Furthermore, we investigate the effects of high pressure on the binding energy of a supersystem consisting of a buckminster fullerene and a corannulene pincer system, and juxtapose various types of surfaces that constitute the boundary between the molecule and the implicit solvent. Our efficient implementation of the GOSTSHYP model paves the way for large-scale simulations of molecules under pressure.

Virtual Human Retina: Simulating Neural Signalling, Degeneration, and Responses to Electrical Stimulation.

Current brain-based visual prostheses pose significant challenges impeding adoption such as the necessarily complex surgeries and occurrence of more substantial side effects due to the sensitivity of the brain. This has led to much effort toward vision restoration being focused on the more approachable part of the brain - the retina. Here we introduce a novel, parameterized simulation platform that enables study of human retinal degeneration and optimization of stimulation strategies. The platform bears immense potential for patient-specific tailoring and serves to enhance artificial vision solutions for individuals with visual impairments. Our virtual retina is developed using the software package, NEURON. This virtual retina platform supports large-scale simulations of over 10,000 neurons whilst upholding strong biological plausibility with multiple important visual pathways and detailed network properties. The comprehensive three-dimensional model includes photoreceptors, horizontal cells, bipolar cells, amacrine cells, and midget and parasol retinal ganglion cells, with comprehensive network connectivity across various eccentricities (1 mm to 5 mm from the fovea) in the human retina. The model is constructed using electrophysiology, immunohistology, and optical coherence tomography imaging data from healthy and degenerate human retinas. We validated our model by replicating numerous experimental observations from human and primate retina, with a particular focus on retinal degeneration. We simulated interactions between diseased retinas and state-of-the-art retinal implants, shedding light on the limitations of commercial retinal prostheses. Our results suggested that appropriate stimulation settings with intraretinal prototype devices could leverage network-mediated activation to achieve activation mosaics more alike that of the retina's response to natural light, promoting the prospect of more naturalistic vision. Our study additionally highlights the importance of controlling inhibitory circuits in the retinal network to induce functionally relevant retinal activity. This study demonstrates the potential of this software package and highlights its utility as a valuable tool for engineers, scientists, and clinicians in the design and optimisation of retinal stimulation devices for both research and educational applications.

Estimation of Monthly Water Table Depth Anomalies Based on the Integration of GRACE and ERA5-Land with Large-Scale Simulations Using Random Forest and LSTM Networks

Estimation of Monthly Water Table Depth Anomalies Based on the Integration of GRACE and ERA5-Land with Large-Scale Simulations Using Random Forest and LSTM Networks

Spontaneous curvature in two-dimensional van der Waals heterostructures

Two-dimensional (2D) van der Waals heterostructures consist of different 2D crystals with diverse properties, constituting the cornerstone of the new generation of 2D electronic devices. Yet interfaces in heterostructures inevitably break bulk symmetry and structural continuity, resulting in delicate atomic rearrangements and novel electronic structures. In this paper, we predict that 2D interfaces undergo “spontaneous curvature”, which means when two flat 2D layers approach each other, they inevitably experience out-of-plane curvature. Based on deep-learning-assisted large-scale molecular dynamics simulations, we observe significant out-of-plane displacements up to 3.8 Å in graphene/BN bilayers induced by curvature, producing a stable hexagonal moiré pattern, which agrees well with experimentally observations. Additionally, the out-of-plane flexibility of 2D crystals enables the propagation of curvature throughout the system, thereby influencing the mechanical properties of the heterostructure. These findings offer fundamental insights into the atomic structure in 2D van der Waals heterostructures and pave the way for their applications in devices.

Research on vehicle scheduling for forest fires in the northern Greater Khingan Mountains

In the face of forest fire emergencies, fast and efficient dispatching of rescue vehicles is an important means of mitigating the damage caused by forest fires, and is an effective method of avoiding secondary damage caused by forest fires, minimizing the damage caused by forest fires to the ecosystem, and mitigating the losses caused by economic development. this paper takes the actual problem as the starting point, constructs a reasonable mathematical model of the problem, for the special characteristics of the emergency rescue vehicle scheduling problem of forest fires, taking into account the actual road conditions in the northern pristine forest area, through the analysis of the cost of paths between the forest area and the highway, to obtain the least obstructed rescue paths, to narrow the gap between the theoretical model and the problem of the actual. Improvement of ordinary genetic algorithm, design of double population strategy selection operation, the introduction of chaotic search initialization population, to improve the algorithm’s solution efficiency and accuracy, through the northern pristine forest area of Daxing’anling real forest fire cases and generation of large-scale random fire point simulation experimental test to verify the effectiveness of the algorithm, to ensure that the effectiveness and reasonableness of the solution to the problem of forest fire emergency rescue vehicle scheduling program. It enriches the solution method of forest fire emergency rescue vehicle dispatching problem in Great Khingan area, which is of great significance to improve the emergency rescue capability in case of sudden forest fire. Through simulation experiments, the proposed Improved Genetic Algorithm (IGA) achieved an average rescue time reduction of 8.5% compared to conventional Genetic Algorithm (GA) and 3.5% compared to Improved Artificial Bee Colony (IABC) algorithm, with an average solution time of 9.4 ms.

Computational Generation of Long-range Axonal Morphologies

Long-range axons are fundamental to brain connectivity and functional organization, enabling communication between different brain regions. Recent advances in experimental techniques have yielded a substantial number of whole-brain axonal reconstructions. While previous computational generative models of neurons have predominantly focused on dendrites, generating realistic axonal morphologies is more challenging due to their distinct targeting. In this study, we present a novel algorithm for axon synthesis that combines algebraic topology with the Steiner tree algorithm, an extension of the minimum spanning tree, to generate both the local and long-range compartments of axons. We demonstrate that our computationally generated axons closely replicate experimental data in terms of their morphological properties. This approach enables the generation of biologically accurate long-range axons that span large distances and connect multiple brain regions, advancing the digital reconstruction of the brain. Ultimately, our approach opens up new possibilities for large-scale in-silico simulations, advancing research into brain function and disorders.