Simulation Of Large Systems Research Articles

Methodologies for the analysis of instantaneous lipid diffusion in MD simulations of large membrane systems.

Interactions between lipids and membrane proteins play a key role in determining the nanoscale dynamic and structural properties of biological membranes. Molecular dynamics (MD) simulations provide a valuable tool for studying membrane models, complementing experimental approaches. It is now possible to simulate large membrane systems, such as simplified models of bacterial and viral envelope membranes. Consequently, there is a pressing need to develop tools to visualize and quantify the dynamics of these immense systems, which typically comprise millions of particles. To tackle this issue, we have developed visual and quantitative analyses of molecular positions and their velocity field using path line, vector field and streamline techniques. This allows us to highlight large, transient flow-like movements of lipids and to better understand crowding within the lipid bilayer. The current study focuses on visualization and analysis of lipid dynamics. However, the methods are flexible and can be readily applied to e.g. proteins and nanoparticles within large complex membranes. The protocols developed here are readily accessible both as a plugin for the molecular visualization program VMD and as a module for the MDAnalysis library.

Graphene mechanics: I. Efficient first principles based Morse potential.

We present a computationally efficient pairwise potential for use in molecular dynamics simulations of large graphene or carbon nanotube systems, in particular, for those under mechanical deformation, and also for mixed systems including biomolecules. Based on the Morse potential, it is only slightly more complex and computationally expensive than a harmonic bond potential, allowing such large or mixed simulations to reach experimentally relevant time scales. By fitting to data obtained from quantum mechanics (QM) calculations to represent bond breaking in graphene patches, we obtain a dissociation energy of 805 kJ mol(-1) which reflects the steepness of the QM potential up to the inflection point. A distinctive feature of our potential is its truncation at the inflection point, allowing a realistic treatment of ruptured C-C bonds without relying on a bond order model. The results obtained from equilibrium MD simulations using our potential compare favorably with results obtained from experiments and from similar simulations with more complex and computationally expensive potentials.

MODEL REDUCTION WITH GEOMETRIC STIFFENING NONLINEARITIES FOR DYNAMIC SIMULATIONS OF MULTIBODY SYSTEMS

This work investigates the implementation of nonlinear model reduction to flexible multibody dynamics. Linear elastic theory will lead to instability issues with rotating beam-like structures, due to the neglecting of the membrane-bending coupling on the beam cross-section. During the past decade, considerable efforts have been focused on the derivation of geometric nonlinear formulation based on nodal coordinates. In order to reduce the computation cost in flexible multibody dynamics, which is extremely important for complex large system simulations, modal reduction is usually implemented to a rotating flexible structure with geometric nonlinearities retained in the model. In this work, a standard model reduction process based on matrix operation is developed, and the essential geometric stiffening nonlinearities are retained in the reduced model. The time responses of a tip point on a rotating Euler–Bernoulli blade are calculated based on two nonlinear reduced models. The first reduced model is derived by the standard matrix operation from a full finite element model and the second reduced model is obtained via the Galerkin method. The matrix operation model reduction process is validated through the comparison of the simulation results obtained from these two different reduced models. An interesting phenomenon is observed in this work: In the nonlinear model, if only quadratic geometric stiffing term is retained, the two reduced models converge to the full finite element model with only one bending mode and two axial modes. While if both quadratic and cubic geometric stiffing terms are retained in the nonlinear equation, the modal-based reduced model will not converge to the finite element model unless all eigenmodes are retained, that is the reduced model has no degree of freedom reduction at all.

Minimal spanning trees at the percolation threshold: A numerical calculation

The fractal dimension of minimal spanning trees on percolation clusters is estimated for dimensions d up to d=5. A robust analysis technique is developed for correlated data, as seen in such trees. This should be a robust method suitable for analyzing a wide array of randomly generated fractal structures. The trees analyzed using these techniques are built using a combination of Prim's and Kruskal's algorithms for finding minimal spanning trees. This combination reduces memory usage and allows for simulation of larger systems than would otherwise be possible. The path length fractal dimension d_{s} of MSTs on critical percolation clusters is found to be compatible with the predictions of the perturbation expansion developed by T. S. Jackson and N. Read [Phys. Rev. E 81, 021131 (2010)].

September 2013 - This Month in JoVE

Here's a look at what's coming up in the November 2013 issue of JoVE: The Journal of Visualized Experiments. In the biomedical sciences, many principles are based on a handful of model organisms—but experimental findings in traditional models don't always translate to humans or other organisms, nor can they answer all fundamental questions. Therefore, scientists often look to non-traditional models, especially in emerging and multidisciplinary fields. In JoVE Neuroscience, we have two video articles with common animals that are not so common in the lab. One is the freshwater leech, the bloodsucking worm that is firmly embedded in traditional medicine and also appreciated in neuroscience. It has an extremely simple nervous system, which allows neuroscientists to study individual neurons and how their connections influence behavior. Titlow et al. demonstrate how to dissect ventral ganglions and culture isolated neurons (which can retain their electrical characteristics for up to 45 days) and obtain stable intracellular recordings in sensory cells. They might not be very appealing, but leeches are valuable tools for teaching neural anatomy, the principles of intracellular recordings, and intercellular communication. Another article in JoVE Neuroscience features the American cockroach, an unwelcome guest in households all over the world. Anyone who has seen cockroaches in action knows how quickly they can change directions, especially when running away from predators. In fact, of all animals, the cockroach has one of the fastest escape responses, which has been attributed to a reflex circuit that involves two horn-like structures called cerci on the abdomen. Hundreds of tiny sensory hairs on the cerci transform stimuli into neural events, which result in escape behavior. Titlow et al. demonstrate how to prepare the cockroach for electrophysiology and perform neural circuit recordings from intact cockroach nervous systems. These techniques can help identify the neural circuits that underlie certain behavioral responses. Our featured article from Clinical and Translational Medicine deals with acute ischemic stroke, where any delay with emergency treatment can negatively affect the outcome. Therefore, Ebinger et al. present the STEMO, or stroke emergency mobile, a specialized ambulance equipped with a computed tomography scanner and point-of-care lab. This allows vital diagnostics and interventions before the patient gets to the hospital. Preliminary examinations are performed and blood samples taken to determine international normalized ratio (INR), which guides anticoagulation treatment. Advanced diagnostics are also possible through telemedicine, where CT images are transmitted electronically to a neuroradiologist on call. If appropriate, thrombolysis is initiated using recombinant tissue plasminogen activator. The emergency room is notified before transport so everything is in place when the patient arrives. While this pre-hospital concept has not yet been tested in all urban and rural settings, it represents a logical approach for shortening time to treatment for stroke patients. Many video articles in JoVE show how to create small-scale simulations of larger systems. This month in JoVE Applied Physics, a very large system is approximated in the lab: the planet Earth, and simulation of its interior differentiation processes to study planetary core formation. Fei does this by mixing olivine silicate, iron powder, and iron sulfide, then loading into a high-pressure cell assembly and heating to 1800°C. The sample is recovered and prepared for 3D imaging. By analyzing high-resolution images and the chemical compositions of the liquid and solid phases, scientists can better understand the process of planetary core formation. You've just had a sneak peek of a few highlights from the November 2013 issue of JoVE. Visit the website to see the full-length articles, plus many more, in JoVE: The Journal of Visualized Experiments.

Test and Analysis GPU-Accelerated in Molecular Dynamics Simulation

GPU computing is the use of a graphics processing unit together with a CPU to accelerate large scale scientific and engineering applications, such as molecule simulation. The paper use NVIDIA Tesla C2050NVIDIA GTX580 and NAMD 2.9 simulates three differences molecule systems: Beta2,SET9 and Ubiquitin. We compared and analyzed the results of the simulations experiment, and come to conclusion that the difference molecule systems will get the difference speed accelerated. The computing times of four GPU is nearly half of the time used by one GPU; and this is especially in the case of macromolecules system. Furthermore, from the GPUs memory utilization rate, the larger the protein system is, the higher the memory use of the GPU is. The performance of NVIDIA GTX580 is only half of the NVIDIAC2050. NVIDIA Tesla C2050 is can satisfy an even larger system simulation.

A hybrid particle–field molecular dynamics approach: a route toward efficient coarse-grained models for biomembranes

This paper gives an overview of the coarse-grained models of phospholipids recently developed by the authors in the frame of a hybrid particle–field molecular dynamics technique. This technique employs a special class of coarse-grained models that are gaining popularity because they allow simulations of large scale systems and, at the same time, they provide sufficiently detailed chemistry for the mapping scheme adopted. The comparison of the computational costs of our approach with standard molecular dynamics simulations is a function of the system size and the number of processors employed in the parallel calculations. Due to the low amount of data exchange, the larger the number of processors, the better are the performances of the hybrid particle–field models. This feature makes these models very promising ones in the exploration of several problems in biophysics.

A Lifetime Estimation Technique for Voltage Source Inverters

This paper presents a method to estimate the inverter lifetime so that we can predict a failure prior to it actually happening. The key contribution of this study is to link the physics of the power devices to a large scale system simulation within a reasonable framework of time. By configuring this technique to a real system, it can be used as a converter design tool or online lifetime estimation tool. In this paper, the presented method is applied to the grid side inverter to show its validity. A power cycling test is designed to gather the lifetime data of a selected insulated gate bipolar transistor (IGBT) module (SKM50GB123D). Die-attach solder fatigue is found out to be the dominant failure mode of this IGBT module under the designed accelerated tests. Furthermore, the crack initiation is found to be highly stress dependent while the crack propagation is almost independent with stress level. Two different damage accumulation methods are used and the estimation results are compared.

Quantum Monte Carlo for large chemical systems: Implementing efficient strategies for petascale platforms and beyond

Various strategies to implement efficiently quantum Monte Carlo (QMC) simulations for large chemical systems are presented. These include: (i) the introduction of an efficient algorithm to calculate the computationally expensive Slater matrices. This novel scheme is based on the use of the highly localized character of atomic Gaussian basis functions (not the molecular orbitals as usually done), (ii) the possibility of keeping the memory footprint minimal, (iii) the important enhancement of single-core performance when efficient optimization tools are used, and (iv) the definition of a universal, dynamic, fault-tolerant, and load-balanced framework adapted to all kinds of computational platforms (massively parallel machines, clusters, or distributed grids). These strategies have been implemented in the QMC=Chem code developed at Toulouse and illustrated with numerical applications on small peptides of increasing sizes (158, 434, 1056, and 1731 electrons). Using 10-80 k computing cores of the Curie machine (GENCI-TGCC-CEA, France), QMC=Chem has been shown to be capable of running at the petascale level, thus demonstrating that for this machine a large part of the peak performance can be achieved. Implementation of large-scale QMC simulations for future exascale platforms with a comparable level of efficiency is expected to be feasible.

<I>“PEM Fuel Cells with Bio-Ethanol Processor Systems: A Multidisciplinary Study of Modelling, Simulation, Fault Diagnosis and Advanced Control”</I>

School of Chemistry, University of Southampton, Highfi eld, Southampton SO17 1BJ, UK Email: l.calvillo@soton.ac.uk Introduction “PEM Fuel Cells with Bio-Ethanol Processor Systems”, edited by Marta S. Basualdo and Diego Feroldi (French-Argentine International Center of Information and Systems Sciences, Argentina) and Rachid Outbib (LSIS-Domaine Universitaire de Saint-Jerome, Marseille, France), who are experts in simulation, control and fault detection of large systems, is part of the series Green Energy and Technology. This book presents some typical problems encountered in proton exchange membrane fuel cell (PEMFC) systems and proposes alternative ways of operation using conventional and advanced control techniques. The book is divided into two main parts. The fi rst part consists of eight chapters and is dedicated to fuel cell systems in terms of modelling, simulation, advanced control and diagnosis; the second part consists of six chapters and deals with fuel processor systems for the production of hydrogen from bioethanol. The writing style is clear and the authors give a detailed description of the systems and processes at the beginning of each chapter, which allows even the non-specialist reader to easily follow the concepts under discussion. However, in some of the chapters, prior knowledge of the specifi c software packages used is needed.

Interaction between hydrogen flux and carbon monolayer on SiC(0001): graphene formation kinetics

Manipulation of graphene-based systems is a formidable challenge, since it requires the control of atomic interactions over long timescales. Although the effectiveness of a certain number of processes has been experimentally demonstrated, the underlying atomic mechanisms are often not understood. An import class of techniques relies on the interaction between hydrogen and graphene, which is the focus of this research. In particular, the growth of epitaxial graphene on SiC(0001) is subject to a single-atom-thick interface carbon layer strongly bound to the substrate, which can be detached through hydrogen intercalation. Here we report that a nucleation phenomenon induces the transformation of this buffer layer into graphene. We study the graphenization dynamics by an ab initio based method that permits the simulation of large systems with an atomic resolution, spanning the time scales from nanoseconds to hours. The early evolution stage (∼ms time scale) is characterised by the formation of a metastable H layer deposited on the C surface. H penetration in the interface between the C monolayer and the SiC(0001) surface is a rare event due to the large penetration barrier, which is ∼2 eV. However, at high H densities, energetically favoured Si-H bonding appears on the substrate's surface. The local increase of the H density at the interface due to statistical transitions leads to the graphenization of the overlying C atoms. Thermally activated density fluctuations promote the formation of these graphene-like islands on the buffer layer: this nucleation phenomenon is evidenced by our simulations at a later evolution stage (>10(2) s at 700 °C for ∼3.6 × 10(15) at. cm(-2) s(-1) H flux). Such nuclei grow and quasi-freestanding graphene forms if the exposition to the H flux continues for a sufficiently long time (∼30 min for the same conditions). We have systematically explored this phenomenon by varying the substrate temperature and the H flux, demonstrating that the surface morphology during graphenization and post-graphenization anneals significantly depends on these variables. The computational findings are consistent with the experimental analyses reported so far and could serve as guidelines for future experimental works on graphene manipulation.

Linear Response Theory and Electronic Transition Energies for a Fully Polarizable QM/Classical Hamiltonian.

A fully polarizable quantum/classical Hamiltonian including SCF (HF or DFT), fluctuating charge, and polarizable continuum regions is introduced and implemented for electronic energies of ground and excited states, using, in the latter case, a linear response formulation. After calibration and validation of the approach, preliminary results are presented for pyrimidine in aqueous solution and for retinal in a rhodopsin mimic. The results are consistent with more tested methodologies and pave the route toward fully consistent yet effective simulations of large systems of technological and/or biological interest in their natural environments.

Analysis of Voltage Profile Problems Due to the Penetration of Distributed Generation in Low-Voltage Secondary Distribution Networks

This paper presents a comprehensive analysis of the possible impacts of different penetration levels of distributed generation (DG) on voltage profiles in low-voltage secondary distribution networks. Detailed models of all system components are utilized in a study that performs hundreds of time-domain simulations of large networked distribution systems using the Electromagnetic Transients Program (EMTP). DGs are allocated in a probabilistic fashion to account for the uncertainties of future installations. The main contribution of this paper is the determination of the maximum amount of DG that secondary distribution networks can withstand without exhibiting undervoltage and overvoltage problems or unexpected load disconnections. This information is important for network planning engineers to facilitate the extension of the maximum penetration limit. The results show that depending on the location, type, and size of the installed DGs, small amounts of DG may cause overvoltage problems. However, large amounts of DG may not cause any voltage problems when properly selected.

Implementación eficiente de una formulación semirrecursiva para la dinámica de sistemas multicuerpo de gran tamaño

This article shows an efficient implementation of a dynamic semi-recursive formulation for large and complex multibody system simulations, with interesting applications in the automotive field and especially with industrial vehicles. These systems tend to have a huge amount of kinematic constraints, becoming usual the presence of redundant but compatible systems of equations. The maths involved in the solution of these problems have a high computational cost, making very challenging to achieve real-time simulations.In this article, two implementations to increase the efficiency of these computations will be shown. The difference between them is the way they consider the Jacobian matrix of the constraint equations. The first one treats this matrix as a dense one, using the BLAS functions to solve the system of equations. The second one takes into account the sparse pattern of the Jacobian matrix, introducing the sparse function MA48 from Harwell.Both methodologies have been applied on two multibody system models with different sizes. The first model is a vehicle IVECO DAILY 35C15 with 17 degrees of freedom. The second one is a semi-trailer truck with 40 degrees of freedom. Taking as a reference the standard C/C++ implementation, the efficiency improvements that have been achieved using dense matrices (BLAS) have been of 15% and 50% respectively. The results in the first model have not improved significantly by using sparse matrices, but in the second one, the times with sparse matrices have been reduced 8% with respect to the BLAS ones.

Low Power Sparse Approximation on Reconfigurable Analog Hardware

Compressed sensing is an important application in signal and image processing which requires solving nonlinear optimization problems. A Hopfield-Network-like analog system is proposed as a solution, using the locally competitive algorithm (LCA) to solve an overcomplete l1 sparse approximation problem. A scalable system architecture using sub-threshold currents is described, including vector matrix multipliers (VMMs) and a nonlinear thresholder. A 4 × 6 nonlinear system is implemented on the RASP 2.9v chip, a field programmable analog array with directly programmable floating gate elements, allowing highly accurate VMMs. The circuit successfully reproduced the outputs of a digital optimization program, converging to within 4.8% rms, and an objective value only 1.3% higher on average. The active circuit consumed 29 μA of current at 2.4 V, and converges on solutions in 240 μs. A smaller 2 × 3 system is also implemented. Extrapolating the scaling trends to a N=1000 node system, the analog LCA compares favorably with state-of-the-art digital solutions, using a small fraction of the power to arrive at solutions ten times faster. Finally, we provide simulations of large scale systems to show the behavior of the system scaled to nontrivial problem sizes.

A multiscale component mode synthesis approach for dynamic analysis of nanostructures

SUMMARYA component mode synthesis‐based multiscale approach is developed for dynamic analysis of nanostructures. The multiscale approach decomposes a nanostructure into atomistic and continuum regions and employs vibrational modes to connect the regions of different scales, enabling a reflectionless atomistic‐to‐continuum coupling. Dynamic response of the coupled atomistic and continuum regions is computed concurrently using a common time scale. Numerical results indicate that the multiscale approach has significant condensation and scaling advantages, and it is well suited for modeling and simulation of large and complex systems. Copyright © 2012 John Wiley & Sons, Ltd.

Large two dimensional Coulomb crystals in a radio frequency surface ion trap

We designed and operated a surface ion trap with an ion-substrate distance of 500 μm, realized with standard printed-circuit-board techniques. The trap has been loaded with up to a few thousand Sr+ ions in the Coulomb-crystal regime. An analytical model of the pseudo-potential allowed us to determine the parameters that drive the trap into anisotropic regimes in which we obtain large (N>150) purely two dimensional (2D) ion Coulomb crystals. These crystals may open a simple and reliable way to experiments on quantum simulations of large 2D systems.

Theoretical Investigation of Hydrogen Adsorption and Dissociation on Iron and Iron Carbide Surfaces Using the ReaxFF Reactive Force Field Method

We have developed a ReaxFF reactive force field to describe hydrogen adsorption and dissociation on iron and iron carbide surfaces relevant for simulation of Fischer–Tropsch (FT) synthesis on iron catalysts. This force field enables large system (>>1000 atoms) simulations of hydrogen related reactions with iron. The ReaxFF force field parameters are trained against a substantial amount of structural and energetic data including the equations of state and heats of formation of iron and iron carbide related materials, as well as hydrogen interaction with iron surfaces and different phases of bulk iron. We have validated the accuracy and applicability of ReaxFF force field by carrying out molecular dynamics simulations of hydrogen adsorption, dissociation and recombination on iron and iron carbide surfaces. The barriers and reaction energies for molecular dissociation on these two types of surfaces have been compared and the effect of subsurface carbon on hydrogen interaction with iron surface is evaluated. We found that existence of carbon atoms at subsurface iron sites tends to increase the hydrogen dissociation energy barrier on the surface, and also makes the corresponding hydrogen dissociative state relatively more stable compared to that on bare iron. These properties of iron carbide will affect the dissociation rate of H2 and will retain more surface hydride species, thus influencing the dynamics of the FT synthesis process.

An efficient parallel dynamics algorithm for simulation of large articulated robotic systems

This paper presents a new parallel computer algorithm for the simulation of large articulated robotic systems. The method employs the Divide and Conquer Algorithm (DCA) multibody methodology and achieves significant increases in speed by using a variation of the Articulated Body Algorithm (ABA) to efficiently construct the DCA subsystems. The method outperforms contemporary parallel dynamics algorithms for a serial topology and degenerates to the ABA in the absence of parallel resources. The implementation and comparison with the ABA demonstrate useful speed increases with as few as two parallel processors and speed ups by a factor of 2.25 in the presence of four processors for serial manipulators. The paper also addresses the practical limitations on obtainable speedups due to the interprocessor communication costs and its consequences on choosing optimal DCA subsystems.

A Fast Finite-Difference Method for Micromagnetics Using the Magnetic Scalar Potential

We propose a method for the stray-field computation of ferromagnetic microstructures via the magnetic scalar potential. The scalar potential is computed using the convolution theorem and the fast Fourier transform. For the discrete convolution an analytical expression for the scalar potential of a uniformly magnetized cuboid is presented. A performance gain of up to 55% compared to common simulation codes is achieved and the memory consumption is reduced by 30%. Since the stray-field computation is the most time consuming part of micromagnetic simulations, this performance gain strongly influences the overall performance. The low memory consumption allows simulations with a high number of simulation cells. This enables simulations of large systems like arrays of coupled magnetic vortices or simulations with high spatial resolution. In conjunction with modern hardware, simulations of microstructures with atomic resolution become feasible.