Search results for "NEURAL NETWORK"

showing 10 items of 1385 documents

Coarsened Lattice Model for Random Granular Systems

1998

In random systems consisting of grains with size distributions the transport properties are difficult to explore by network models. However, the concentration dependence of effective conductivity and its critical properties can be considered within coarsened lattice model proposed that takes into account information from experimentally known size histograms. For certain classes of size distributions the specific local arrangements of grains can induce either symmetrical or unsymmetrical critical behaviour at two threshold concentrations. Using histogram related parameters the non-monotonic behaviour of the conductor-insulator and conductor-superconductor threshold is demonstrated.

Materials scienceStatistical Mechanics (cond-mat.stat-mech)Critical phenomenaFOS: Physical sciencesDisordered Systems and Neural Networks (cond-mat.dis-nn)ConductivityCondensed Matter - Disordered Systems and Neural NetworksCondensed Matter PhysicsGrain sizeElectronic Optical and Magnetic MaterialsDistribution functionPercolationHistogramStatistical physicsLattice model (physics)Condensed Matter - Statistical MechanicsNetwork model
researchProduct

Slow dynamics in ion-conducting sodium silicate melts: Simulation and mode-coupling theory

2005

A combination of molecular-dynamics (MD) computer simulation and mode-coupling theory (MCT) is used to elucidate the structure-dynamics relation in sodium-silicate melts (NSx) of varying sodium concentration. Using only the partial static structure factors from the MD as an input, MCT reproduces the large separation in relaxation time scales of the sodium and the silicon/oxygen components. This confirms the idea of sodium diffusion channels which are reflected by a prepeak in the static structure factors around 0.95 A^-1, and shows that it is possible to explain the fast sodium-ion dynamics peculiar to these mixtures using a microscopic theory.

Materials scienceStatistical Mechanics (cond-mat.stat-mech)SiliconSodiumFOS: Physical sciencesGeneral Physics and Astronomychemistry.chemical_elementSodium silicateDisordered Systems and Neural Networks (cond-mat.dis-nn)Condensed Matter - Disordered Systems and Neural NetworksOxygenIonchemistry.chemical_compoundchemistryChemical physicsMode couplingDiffusion (business)Microscopic theoryCondensed Matter - Statistical MechanicsEurophysics Letters (EPL)
researchProduct

Light-induced ionic processes in optical oxide glasses

1991

Abstract The density of optical glasses is changed by the influence of light capable of generating color centers in these materials. Such defect generation is not only an electronic process, but an atomic displacement is also necessary. The strong localization of electronic and vibrational excitations in the glass network leads to the high efficiency of sub-threshold defect generation. Secondary ionic processes lead to the changes of basic glass properties (light refractive index, density, mechanical strength, etc.); thus, it is possible to use optical glasses as light detectors for appropriate wavelengths.

Materials sciencebusiness.industryDetectorOxideIonic bondingCondensed Matter PhysicsCondensed Matter::Disordered Systems and Neural NetworksElectronic Optical and Magnetic Materialschemistry.chemical_compoundWavelengthchemistryMechanical strengthMaterials ChemistryCeramics and CompositesLight inducedOptoelectronicsbusinessAtomic displacementRefractive indexJournal of Non-Crystalline Solids
researchProduct

Statics and dynamics of colloid-polymer mixtures near their critical point of phase separation: A computer simulation study of a continuous Asakura–O…

2008

We propose a new coarse-grained model for the description of liquid-vapor phase separation of colloid-polymer mixtures. The hard-sphere repulsion between colloids and between colloids and polymers, which is used in the well-known Asakura-Oosawa (AO) model, is replaced by Weeks-Chandler-Anderson potentials. Similarly, a soft potential of height comparable to thermal energy is used for the polymer-polymer interaction, rather than treating polymers as ideal gas particles. It is shown by grand-canonical Monte Carlo simulations that this model leads to a coexistence curve that almost coincides with that of the AO model and the Ising critical behavior of static quantities is reproduced. Then the …

Materials sciencecritical pointsMonte Carlo methodFOS: Physical sciencesGeneral Physics and AstronomyThermodynamicsCondensed Matter - Soft Condensed MatterCritical point (mathematics)Molecular dynamicscolloidspolymer solutionsPhysical and Theoretical Chemistryliquid-vapour transformationsBinodalliquid mixturesLennard-Jones potentialMonte Carlo methodsDisordered Systems and Neural Networks (cond-mat.dis-nn)Statistical mechanicsCondensed Matter - Disordered Systems and Neural Networksself-diffusionIdeal gasliquid theoryCondensed Matter::Soft Condensed Mattermolecular dynamics methodLennard-Jones potentialSoft Condensed Matter (cond-mat.soft)Ising modelstatistical mechanicsphase separationThe Journal of Chemical Physics
researchProduct

Effect of mixing and spatial dimension on the glass transition

2009

We study the influence of composition changes on the glass transition of binary hard disc and hard sphere mixtures in the framework of mode coupling theory. We derive a general expression for the slope of a glass transition line. Applied to the binary mixture in the low concentration limits, this new method allows a fast prediction of some properties of the glass transition lines. The glass transition diagram we find for binary hard discs strongly resembles the random close packing diagram. Compared to 3D from previous studies, the extension of the glass regime due to mixing is much more pronounced in 2D where plasticization only sets in at larger size disparities. For small size disparitie…

Materials sciencepacs:82.70.DdCondensed matter physicsStatistical Mechanics (cond-mat.stat-mech)business.industryDiagramRandom close packBinary numberFOS: Physical sciencesCondensed Matter - Soft Condensed MatterCondensed Matter::Disordered Systems and Neural NetworksCondensed Matter::Soft Condensed MatterOpticsPhase (matter)Mode couplingSoft Condensed Matter (cond-mat.soft)ddc:530Glass transitionbusinesspacs:64.70.Q-Mixing (physics)Condensed Matter - Statistical Mechanicspacs:64.70.PLine (formation)
researchProduct

Soft Sensor Transferability between Lines of a Sulfur Recovery Unit

2021

Abstract Soft Sensors (SSs) are mathematical models that allow real-time estimation of hard-to-measure variables as a function of easy-to-measure ones in an industrial process, emulating the behavior of existing sensors when they are, for instance, taken off for maintenance. The Sulfur Recovery Unit (SRU) from a refinery is taken in exam. Recurrent Neural Networks (RNN) can capture the nonlinearity of such process but present a high complexity training and a very time-consuming structure optimization. For this reason, strategies to use pre-existing models are here examined by testing the transferability of the SSs between two parallel lines of the process.

Mathematical modelComputer sciencemedia_common.quotation_subjectProcess (computing)transferable soft sensor; nonlinear model; recurrent neural network; monitoring; prediction; inferential modelControl engineeringpredictionSoft sensorParallelRefineryNonlinear systemmonitoringRecurrent neural networkinferential modelControl and Systems Engineeringnonlinear modelrecurrent neural networkFunction (engineering)media_commontransferable soft sensor
researchProduct

Least-squares temporal difference learning based on an extreme learning machine

2014

Abstract Reinforcement learning (RL) is a general class of algorithms for solving decision-making problems, which are usually modeled using the Markov decision process (MDP) framework. RL can find exact solutions only when the MDP state space is discrete and small enough. Due to the fact that many real-world problems are described by continuous variables, approximation is essential in practical applications of RL. This paper is focused on learning the value function of a fixed policy in continuous MPDs. This is an important subproblem of several RL algorithms. We propose a least-squares temporal difference (LSTD) algorithm based on the extreme learning machine. LSTD is typically combined wi…

Mathematical optimizationArtificial neural networkArtificial IntelligenceCognitive NeuroscienceBellman equationReinforcement learningState spaceMarkov decision processTemporal difference learningComputer Science ApplicationsMathematicsExtreme learning machineCurse of dimensionalityNeurocomputing
researchProduct

Fast Convergence of Neural Networks by Application of a New Min-Max Algorithm

1992

Abstract The paper presents a new application of the min-max method, an original algorithm previously successfully applied in other areas and based on a combination of the quasi-Newton and steepest descent methods in order to find the weights minimising the error function of a feed forward neural networks. Preliminary results, obtained by applying the proposed method to a simple 2-2-1 architecture on small Boolean learning problems, are very promising.

Mathematical optimizationError functionArtificial neural networkComputer scienceSimple (abstract algebra)Convergence (routing)MinimaxGradient descent
researchProduct

A New Min-Max Optimisation Approach for Fast Learning Convergence of Feed-Forward Neural Networks

1993

One of the most critical aspect for a wide use of neural networks to real world problems is related to the learning process which is known to be computational expensive and time consuming.

Mathematical optimizationError functionArtificial neural networkWake-sleep algorithmComputer sciencebusiness.industryConvergence (routing)Process (computing)Feed forward neuralArtificial intelligenceDescent directionbusinessGeneralization error
researchProduct

Conflict resolution in the multi-stakeholder stepped spillway design under uncertainty by machine learning techniques

2021

Abstract The optimal spillway design is of great significance since these structures can reduce erosion downstream of the dams. This study proposes a risk-based optimization framework for a stepped spillway to achieve an economical design scenario with the minimum loss in hydraulic performance. Accordingly, the stepped spillway was simulated in the FLOW-3D® model, and the validated model was repeatedly performed for various geometric states. The results were used to form a Multilayer Perceptron artificial neural network (MLP-ANN) surrogate model. Then, a risk-based optimization model was formed by coupling the MLP-ANN and NSGA-II. The concept of conditional value at risk (CVaR) was utilized…

Mathematical optimizationExpected shortfallSpillwaySurrogate modelArtificial neural networkComputer scienceCVARMultilayer perceptronConflict resolutionStepped spillwayVDP::Technology: 500::Information and communication technology: 550SoftwareApplied Soft Computing
researchProduct