Network Simulations Research Articles

Flows Over Time as Continuous Limits of Packet-Based Network Simulations

This study examines the connection between an agent-based transport simulation and Nash flows over time. While the former is able to represent many details of traffic and model large-scale, real-world traffic situations with a co-evolutionary approach, the latter provides an environment for provable mathematical statements and results on exact user equilibria. The flow dynamics of both models are very similar with the main difference that the simulation is discrete in terms of vehicles and time while the flows over time model considers continuous flows and continuous time. This raises the question whether Nash flows over time are the limit of the convergence process when decreasing the vehicle and time step size in the simulation coherently. The experiments presented in this study indicate this strong connection which provides a justification for the analytical model and a theoretical foundation for the simulation.

Layer two compression for erroneous wireless

Wireless connection such as 802.11 and cellar networks is preferred for internet connections as its quick set up and mobility. However, radio propagations are prone to signal errors, noises and interferences. Even more powerful modulations and coding were developed; they do not change the nature of error sensitive in wireless link. When network link worsens, the available bandwidth decreases. This exerts problems for the user speed requirement. This paper examines layer two compressions on decreasing link quality. The 802.11 network simulations based on Bianchi model and Shannon theory show that compression technique on layer two increases throughput about 10.27%, although delay increases about two times.

A Siamese neural network model for the prioritization of metabolic disorders by integrating real and simulated data.

Untargeted metabolomic approaches hold a great promise as a diagnostic tool for inborn errors of metabolisms (IEMs) in the near future. However, the complexity of the involved data makes its application difficult and time consuming. Computational approaches, such as metabolic network simulations and machine learning, could significantly help to exploit metabolomic data to aid the diagnostic process. While the former suffers from limited predictive accuracy, the latter is normally able to generalize only to IEMs for which sufficient data are available. Here, we propose a hybrid approach that exploits the best of both worlds by building a mapping between simulated and real metabolic data through a novel method based on Siamese neural networks (SNN). The proposed SNN model is able to perform disease prioritization for the metabolic profiles of IEM patients even for diseases that it was not trained to identify. To the best of our knowledge, this has not been attempted before. The developed model is able to significantly outperform a baseline model that relies on metabolic simulations only. The prioritization performances demonstrate the feasibility of the method, suggesting that the integration of metabolic models and data could significantly aid the IEM diagnosis process in the near future. Metabolic datasets used in this study are publicly available from the cited sources. The original data produced in this study, including the trained models and the simulated metabolic profiles, are also publicly available (Messa et al., 2020).

Spin-$\frac{1}{2}$ kagome Heisenberg antiferromagnet with strong breathing anisotropy

We study the zero-temperature phase diagram of the spin-1/2 Heisenberg model with breathing anisotropy (i.e., with different coupling strength on the upward and downward triangles) on the kagome lattice. Our study relies on large scale tensor network simulations based on infinite projected entangled-pair state and infinite projected entangled-simplex state methods adapted to the kagome lattice. Our energy analysis suggests that the U(1) algebraic quantum spin-liquid (QSL) ground-state of the isotropic Heisenberg model is stable up to very large breathing anisotropy until it breaks down to a critical lattice-nematic phase that breaks rotational symmetry in real space through a first-order quantum phase transition. Our results also provide further insight into the recent experiment on vanadium oxyfluoride compounds which has been shown to be relevant platforms for realizing QSL in the presence of breathing anisotropy.

Simulation of a detoxifying organ function: Focus on hemodynamics modeling and convection-reaction numerical simulation in microcirculatory networks.

When modeling a detoxifying organ function, an important component is the impact of flow on the metabolism of a compound of interest carried by the blood. We here study the effects of red blood cells (such as the Fahraeus-Lindqvist effect and plasma skimming) on blood flow in typical microcirculatory components such as tubes, bifurcations and entire networks, with particular emphasis on the liver as important representative of detoxifying organs. In one of the plasma skimming models, under certain conditions, oscillations between states are found and analyzed in a methodical study to identify their causes and influencing parameters. The flow solution obtained is then used to define the velocity at which a compound would be transported. A convection-reaction equation is studied to simulate the transport of a compound in blood and its uptake by the surrounding cells. Different types of signal sharpness have to be handled depending on the application to address different temporal compound concentration profiles. To permit executing the studied models numerically stable and accurate, we here extend existing transport schemes to handle converging bifurcations, and more generally multi-furcations. We study the accuracy of different numerical schemes as well as the effect of reactions and of the network itself on the bolus shape. Even though this study is guided by applications in liver micro-architecture, the proposed methodology is general and can readily be applied to other capillary network geometries, hence to other organs or to bioengineered network designs.

Comprehensive Estimates of Potential Synaptic Connections in Local Circuits of the Rodent Hippocampal Formation by Axonal-Dendritic Overlap.

A quantitative description of the hippocampal formation synaptic architecture is essential for understanding the neural mechanisms of episodic memory. Yet the existing knowledge of connectivity statistics between different neuron types in the rodent hippocampus only captures a mere 5% of this circuitry. We present a systematic pipeline to produce first-approximation estimates for most of the missing information. Leveraging the www.Hippocampome.org knowledge base, we derive local connection parameters between distinct pairs of morphologically identified neuron types based on their axonal-dendritic overlap within every layer and subregion of the hippocampal formation. Specifically, we adapt modern image analysis technology to determine the parcel-specific neurite lengths of every neuron type from representative morphologic reconstructions obtained from either sex. We then compute the average number of synapses per neuron pair using relevant anatomic volumes from the mouse brain atlas and ultrastructurally established interaction distances. Hence, we estimate connection probabilities and number of contacts for >1900 neuron type pairs, increasing the available quantitative assessments more than 11-fold. Connectivity statistics thus remain unknown for only a minority of potential synapses in the hippocampal formation, including those involving long-range (23%) or perisomatic (6%) connections and neuron types without morphologic tracings (7%). The described approach also yields approximate measurements of synaptic distances from the soma along the dendritic and axonal paths, which may affect signal attenuation and delay. Overall, this dataset fills a substantial gap in quantitatively describing hippocampal circuits and provides useful model specifications for biologically realistic neural network simulations, until further direct experimental data become available.SIGNIFICANCE STATEMENT The hippocampal formation is a crucial functional substrate for episodic memory and spatial representation. Characterizing the complex neuron type circuit of this brain region is thus important to understand the cellular mechanisms of learning and navigation. Here we present the first numerical estimates of connection probabilities, numbers of contacts per connected pair, and synaptic distances from the soma along the axonal and dendritic paths, for more than 1900 distinct neuron type pairs throughout the dentate gyrus, CA3, CA2, CA1, subiculum, and entorhinal cortex. This comprehensive dataset, publicly released online at www.Hippocampome.org, constitutes an unprecedented quantification of the majority of the local synaptic circuit for a prominent mammalian neural system and provides an essential foundation for data-driven, anatomically realistic neural network models.

Measurement based parameters estimation of large scale wind farm dynamic equivalent model

For the past few decades, the drastic increase in the installed capacity of wind farms (WFs) has necessitated a computationally efficient dynamic equivalent model of a WF, which can be used for accurate power network simulations. Various doubly-fed induction generators (DFIGs) with sophisticated power control, rotor, and grid side converter control based models for wind turbine (WT) exist in the literature. However, these models are computationally burdensome and restrict their application for modeling large WFs. Thus, an equivalent DFIG based model of a WF, with the rotor circuit modeled as a constant current source, is proposed in this paper. The proposed model is simple and computationally efficient as no rotor measurements are required, and voltage and current measurements at the point of common coupling (PCC) of the WF are used in extended Kalman filter (EKF) to evaluate the dynamic parameters of the equivalent DFIG. A Ybus based reduction is also proposed for the equivalent collector system’s impedance. Simulation results on a test WF in MATLAB Simulink and DIgSILENT prove the proposed model’s efficacy over the existing equivalent models for different operating conditions like wake effect and contingencies in the interconnected power system network.

Bayesian networks: regenerative Gibbs samplings

Gibbs samplings is a Markov Chain Monte Carlo technique for estimating conditional probabilities in Bayesian networks. A major problem of Gibbs sampling is the dependency of the generated chain of samples. Thus the estimates are biased unless the initial value of the chain is drawn from the target distribution. One elegant method to overcome the initial bias is regenerative samplings. We reported elsewhere the “stationary minorization condition” that makes any Markov Chain Monte Carlo technique regenerative. In this paper, we show how this condition can be easily met in the simulations of any Bayesian network.

Multiscale modeling of blood flow to assess neurological complications in patients supported by venoarterial extracorporeal membrane oxygenation

Computational blood flow models in large arteries elucidate valuable relationships between cardiovascular diseases and hemodynamics, leading to improvements in treatment planning and clinical decision making. One such application with potential to benefit from simulation is venoarterial extracorporeal membrane oxygenation (VA-ECMO), a support system for patients with cardiopulmonary failure. VA-ECMO patients develop high rates of neurological complications, partially due to abnormal blood flow throughout the vasculature from the VA-ECMO system. To better understand these hemodynamic changes, it is important to resolve complex local flow parameters derived from three-dimensional (3D) fluid dynamics while also capturing the impact of VA-ECMO support throughout the systemic arterial system. As high-resolution 3D simulations of the arterial network remain computationally expensive and intractable for large studies, a validated, multiscale model is needed to compute both global effects and high-fidelity local hemodynamics. In this work, we developed and demonstrated a framework to model hemodynamics in VA-ECMO patients using coupled 3D and one-dimensional (1D) models (1D→3D). We demonstrated the ability of these multiscale models to simulate complex flow patterns in specific regions of interest while capturing bulk flow throughout the systemic arterial system. We compared 1D, 3D, and 1D→3D coupled models and found that multiscale models were able to sufficiently capture both global and local hemodynamics in the cerebral arteries and aorta in VA-ECMO patients. This study is the first to develop and compare 1D, 3D, and 1D→ 3D coupled models on the larger arterial system scale in VA-ECMO patients, with potential use for other large scale applications.

Scalable Approximate Bayesian Computation for Growing Network Models via Extrapolated and Sampled Summaries.

Approximate Bayesian computation (ABC) is a simulation-based likelihood-free method applicable to both model selection and parameter estimation. ABC parameter estimation requires the ability to forward simulate datasets from a candidate model, but because the sizes of the observed and simulated datasets usually need to match, this can be computationally expensive. Additionally, since ABC inference is based on comparisons of summary statistics computed on the observed and simulated data, using computationally expensive summary statistics can lead to further losses in efficiency. ABC has recently been applied to the family of mechanistic network models, an area that has traditionally lacked tools for inference and model choice. Mechanistic models of network growth repeatedly add nodes to a network until it reaches the size of the observed network, which may be of the order of millions of nodes. With ABC, this process can quickly become computationally prohibitive due to the resource intensive nature of network simulations and evaluation of summary statistics. We propose two methodological developments to enable the use of ABC for inference in models for large growing networks. First, to save time needed for forward simulating model realizations, we propose a procedure to extrapolate (via both least squares and Gaussian processes) summary statistics from small to large networks. Second, to reduce computation time for evaluating summary statistics, we use sample-based rather than census-based summary statistics. We show that the ABC posterior obtained through this approach, which adds two additional layers of approximation to the standard ABC, is similar to a classic ABC posterior. Although we deal with growing network models, both extrapolated summaries and sampled summaries are expected to be relevant in other ABC settings where the data are generated incrementally.

Key drivers of trucking safety climate from the perspective of leader-member exchange: Bayesian network predictive modeling approach

PurposeSafety climate, which is defined as workers’ shared perceptions of organizational policies, procedures, and practices as they relate to the true or relative value and importance of safety within an organization, is one of the best indicators of organizational safety outcomes. This study identifies key drivers of safety climate from the perspective of leader-member exchange (LMX). LMX is a theory describing the nature and processes of social interactions between a supervisor and a subordinate. This study examines the impact of individual drivers and combinations of drivers on safety climate through Bayesian Network simulations to predict practices which most effectively improve safety climate in the trucking industry. MethodSurvey data were collected from 5083 truck drivers in a large U.S. trucking company. Bayesian Network analysis was used to identify key drivers (factors) of safety climate and the best joint strategies for improvement. The impact of the drivers on safety climate was assessed and the simulation identified their potential impact independently and in concert with other drivers. ResultsThe results from Bayesian Network analyses showed that the effects of LMX on organization- and group-level safety climate were conditionally dependent on four other drivers including psychological ownership, supervisory integrity, situation awareness, and safety communication. Among the five contributing factors, supervisory integrity and LMX had the strongest independent effects on organization- and group-level safety climate. Moreover, the results indicated that the best two joint strategies for promoting organizational (company/top management level) safety climate were LMX and psychological ownership as well as LMX and situation awareness, whereas the best two joint strategies for improving group (workgroup/supervisor level) safety climate were joint optimization of LMX and safety communication as well as LMX and psychological ownership. ImplicationsBased on the study results, the strategies that may have the most potential to improve trucking safety climate are: enhancing leaders' ability to engage in high-quality exchanges (e.g., caring about employees), developing training to encourage employees/leaders to deliver on promises, and providing employees with more autonomy to enhance their ownership.

Physiologically Based Precision Dosing Approach for Drug-Drug-Gene Interactions: A Simvastatin Network Analysis.

Drug-drug interactions (DDIs) and drug-gene interactions (DGIs) are well known mediators for adverse drug reactions (ADRs), which are among the leading causes of death in many countries. Because physiologically based pharmacokinetic (PBPK) modeling has demonstrated to be a valuable tool to improve pharmacotherapy affected by DDIs or DGIs, it might also be useful for precision dosing in extensive interaction network scenarios. The presented work proposes a novel approach to extend the prediction capabilities of PBPK modeling to complex drug-drug-gene interaction (DDGI) scenarios. Here, a whole-body PBPK network of simvastatin was established, including three polymorphisms (SLCO1B1 (rs4149056), ABCG2 (rs2231142), and CYP3A5 (rs776746)) and four perpetrator drugs (clarithromycin, gemfibrozil, itraconazole, and rifampicin). Exhaustive network simulations were performed and ranked to optimize 10,368 DDGI scenarios based on an exposure marker cost function. The derived dose recommendations were translated in a digital decision support system, which is available at simvastatin.precisiondosing.de. Although the network covers only a fraction of possible simvastatin DDGIs, it provides guidance on how PBPK modeling could be used to individualize pharmacotherapy in the future. Furthermore, the network model is easily extendable to cover additional DDGIs. Overall, the presented work is a first step toward a vision on comprehensive precision dosing based on PBPK models in daily clinical practice, where it could drastically reduce the risk of ADRs.

Deformation Model of Chains and Networks with Extendable Bonds

The nonlinear stress–strain response of synthetic and biological networks and gels manifested in a monotonic increase of an instantaneous modulus is a result of the nonlinear deformation of individual network strands. To describe this behavior, we develop a nonlinear network deformation model, which relates macroscopic stress–strain response with force-elongation behavior of polymer chains with bending rigidity and extendable bonds. In particular, network mechanical properties such as structural shear modulus, extensibility ratio, and bond deformation modulus are expressed in terms of chain bending constant, Kuhn length, bond elastic constant, and chain sizes in undeformed and fully extended states. The chain deformation model is tested by molecular dynamics simulations of deformation of coarse-grained chains with harmonic and finitely extensible nonlinear elastic (FENE) bonds and is applied to analyze the DNA deformation data. The network model predictions are confirmed by molecular dynamics simulations of diamond networks made of coarse-grained chains and are used to describe experimental data for biological networks of collagen, fibrin, and neurofilaments.

Index system and methods for comprehensive assessment of cross-border power grid interconnection projects

With the global economy integration and progress in energy transformation, it has become a general trend to surpass national boundaries to achieve wider and optimal energy resource allocations. Consequently, there is a critical need to adopt scientific approaches in assessing cross-border power grid interconnection projects. First, considering the promotion of large-scale renewable energy resources and improvements in system adequacy, a comprehensive assessment index system, including costs, socio-economic benefits, environmental benefits, and technical benefits, is established in this study. Second, a synthetic assessment framework is proposed for cross-border power grid interconnection projects based on the index system comprising cost–benefit analysis, with market and network simulations, iterative methods for indicator weight evaluation, and technique for order preference by similarity to an ideal solution (TOPSIS) method for the project rankings. Finally, by assessing and comparing three cross-border projects between Europe and Asia, the proposed index system and assessment framework have been proved to be effective and feasible; the results of this system can thus support investment decision-making related to such projects in the future.

Optimal Configuration of LoRa Networks in Smart Cities

Long range (LoRa) is a wireless communication standard specifically targeted for resource-constrained Internet of Things (IoT) devices. LoRa is a promising solution for smart city applications as it can provide long-range connectivity with a low energy consumption. The number of LoRa-based networks is growing due to its operation in the unlicensed radio bands and the ease of network deployments. However, the scalability of such networks suffers as the number of deployed devices increases. In particular, the network performance drops due to increased contention and interference in the unlicensed LoRa radio bands. This results in an increased number of dropped messages and, therefore, unreliable network communications. Nevertheless, network performance can be improved by appropriately configuring the radio parameters of each node. To this end, in this article we formulate integer linear programming models to configure LoRa nodes with the optimal parameters that allow all devices to reliably send data with a low energy consumption. We evaluate the performance of our solutions through extensive network simulations considering different types of realistic deployments. We find that our solution consistently achieves a higher delivery ratio (up to 8% higher) than the state of the art with minimal energy consumption. Moreover, the higher delivery ratio is achieved by a large percentage of nodes in each network, thereby resulting in a fair allocation of radio resources. Finally, the optimal network configurations are obtained within a short time, usually much faster than the state of the art. Thus, our solution can be readily used by network operators to determine optimal configurations for their IoT deployments, resulting in improved network reliability.

Biological network growth in complex environments: A computational framework.

Spatial biological networks are abundant on all scales of life, from single cells to ecosystems, and perform various important functions including signal transmission and nutrient transport. These biological functions depend on the architecture of the network, which emerges as the result of a dynamic, feedback-driven developmental process. While cell behavior during growth can be genetically encoded, the resulting network structure depends on spatial constraints and tissue architecture. Since network growth is often difficult to observe experimentally, computer simulations can help to understand how local cell behavior determines the resulting network architecture. We present here a computational framework based on directional statistics to model network formation in space and time under arbitrary spatial constraints. Growth is described as a biased correlated random walk where direction and branching depend on the local environmental conditions and constraints, which are presented as 3D multilayer grid. To demonstrate the application of our tool, we perform growth simulations of a dense network between cells and compare the results to experimental data from osteocyte networks in bone. Our generic framework might help to better understand how network patterns depend on spatial constraints, or to identify the biological cause of deviations from healthy network function.

Population downscaling in multi-agent transportation simulations: A review and case study

Simulating the dynamics and evolution of metropolitan transportation systems serving millions of travelers remains a difficult task, beyond existing standard software's abilities. MATSim (Multi-Agent Transportation Simulation) is the only high-resolution spatially-explicit framework that allows intrinsic population downscaling - simulating the entire system's dynamics based only on a fraction k of the traveling population. Till now, the choice of k was dictated by hardware performance, and a common rule was not to downscale below 10%. We investigate downscaling in MATSim by comparing the aggregate and disaggregate statistics that describe the dynamics of car traffic in full-scaled and downscaled simulations of the Sioux Falls test case road network. Simulations with 25% or higher shares of the traveler population preserve all major urban traffic statistics. Within the 10–25% interval, downscaling becomes unstable for some of the statistics. For scenarios that are downscaled 10% and below, statistics can substantially deviate from the full-scale model. We further discuss the problems related to a multimodal transportation simulation's downscaling that also includes public transportation.

A brain-inspired network architecture for cost-efficient object recognition in shallow hierarchical neural networks

The brain successfully performs visual object recognition with a limited number of hierarchical networks that are much shallower than artificial deep neural networks (DNNs) that perform similar tasks. Here, we show that long-range horizontal connections (LRCs), often observed in the visual cortex of mammalian species, enable such a cost-efficient visual object recognition in shallow neural networks. Using simulations of a model hierarchical network with convergent feedforward connections and LRCs, we found that the addition of LRCs to the shallow feedforward network significantly enhances the performance of networks for image classification, to a degree that is comparable to much deeper networks. We found that a combination of sparse LRCs and dense local connections dramatically increases performance per wiring cost. From network pruning with gradient-based optimization, we also confirmed that LRCs could emerge spontaneously by minimizing the total connection length while maintaining performance. Ablation of emerged LRCs led to a significant reduction of classification performance, which implies these LRCs are crucial for performing image classification. Taken together, our findings suggest a brain-inspired strategy for constructing a cost-efficient network architecture to implement parsimonious object recognition under physical constraints such as shallow hierarchical depth.

Learning cell for superconducting neural networks

An energy-efficient adiabatic learning neuro cell is proposed. The cell can be used for on-chip learning of adiabatic superconducting artificial neural networks. The static and dynamic characteristics of the proposed learning cell have been investigated. Optimization of the learning cell parameters was performed within simulations of the multi-layer neural network supervised learning with the resilient propagation method.

From micro-scale to macro-scale modeling of solute transport in drying capillary porous media

Drying of capillary porous media initially saturated with saline water is central to many engineering and environmental applications. In order to predict the evolution of solute concentration in a porous medium, the macroscopic continuum models (CMs) are commonly employed. However, the predictive aptitudes of the CMs have been questioned. In this work, we solve the classical advection-diffusion equation for solute transport in an isothermally drying capillary porous medium for the limiting condition of capillary-dominated regime. The solution of the continuum model is compared with pore network simulations. The results of both models are analyzed in terms of local solute concentration profiles for different values of network saturation. On this basis, the ability of the continuum model to reciprocate the pore network results is assessed. The degree of heterogeneity in the liquid phase structure is characterized by performing pore network Monte Carlo simulations distinguishing the percolating liquid cluster from the non-percolating isolated clusters. Based on the statistical analysis of Monte Carlo simulations, the probability of first solid crystals to appear in the respective liquid phase elements is discussed. We observe that solute enrichment is more pronounced in the isolated single liquid throats and isolated clusters due to lack or significant hindrance to back-diffusion as a result of discontinuity in the liquid phase.