Conditional Generative Adversarial Network Enabled Localized Stress Recovery of Periodic Composites

Structural damage in heterogeneousmaterials typically originates frommicrostructures where stress concentration occurs.Therefore,evaluating the magnitude and location of localized stress distributions within microstructures under external loading is crucial.Repeating unit cells(RUCs)are commonly used to represent microstructural details and homogenize the effective response of composites.This work develops a machine learning-based micromechanics tool to accurately predict the stress distributions of extracted RUCs.The locally exact homogenization theory efficiently generates the microstructural stresses of RUCs with a wide range of parameters,including volume fraction,fiber/matrix property ratio,fiber shapes,and loading direction.Subsequently,the conditional generative adversarial network(cGAN)is employed and constructed as a surrogate model to establish the statistical correlation between these parameters and the corresponding localized stresses.The stresses predicted by cGAN are validated against the remaining true data not used for training,showing good agreement.This work demonstrates that the cGAN-based micromechanics tool effectively captures the local responses of composite RUCs.It can be used for predicting potential crack initiations starting from microstructures and evaluating the effective behavior of periodic composites.

Flexible Polydimethylsiloxane Composite with Multi-Scale Conductive Network for Ultra-Strong Electromagnetic Interference Protection

Highly conductive polymer composites(CPCs) with excellent mechanical flexibility are ideal materials for designing excellent electromagnetic interference(EMI) shielding materials,which can be used for the electromagnetic interference protection of flexible electronic devices.It is extremely urgent to fabricate ultra-strong EMI shielding CPCs with efficient conductive networks.In this paper,a novel silver-plated polylactide short fiber(Ag@PL ASF,AAF) was fabricated and was integrated with carbon nanotubes(CNT) to construct a multi-scale conductive network in polydimethylsiloxane(PDMS) matrix.The multi-scale conductive network endowed the flexible PDMS/AAF/CNT composite with excellent electrical conductivity of 440 S m-1and ultra-strong EMI shielding effectiveness(EMI SE) of up to 113 dB,containing only 5.0 vol% of AAF and 3.0 vol% of CNT(11.1wt% conductive filler content).Due to its excellent flexibility,the composite still showed 94% and 90% retention rates of EMI SE even after subjected to a simulated aging strategy(60℃ for 7 days) and 10,000 bending-releasing cycles.This strategy provides an important guidance for designing excellent EMI shielding materials to protect the workspace,environment and sensitive circuits against radiation for flexible electronic devices.

Effect of land use on soil nematode community composition and co-occurrence network relationship

Land use influences soil biota community composition and diversity,and then belowground ecosystem processes and functions.To characterize the effect of land use on soil biota,soil nematode communities in crop land,forest land and fallow land were investigated in six regions of northern China.Generic richness,diversity,abundance and biomass of soil nematodes was the lowest in crop land.The richness and diversity of soil nematodes were 28.8and 15.1%higher in fallow land than in crop land,respectively.No significant differences in soil nematode indices were found between forest land and fallow land,but their network keystone genera composition was different.Among the keystone genera,50%of forest land genera were omnivores-predators and 36%of fallow land genera were bacterivores.The proportion of fungivores in forest land was 20.8%lower than in fallow land.The network complexity and the stability were lower in crop land than forest land and fallow land.Soil pH,NH_(4)^(+)-N and NO_(3)^(–)-N were the major factors influencing the soil nematode community in crop land while soil organic carbon and moisture were the major factors in forest land.Soil nematode communities in crop land influenced by artificial management practices were more dependent on the soil environment than communities in forest land and fallow land.Land use induced soil environment variation and altered network relationships by influencing trophic group proportions among keystone nematode genera.

Tungsten oxide/nitrogen-doped carbon nanotubes composite catalysts for enhanced redox kinetics in lithium-sulfur batteries

The sluggish redox kinetics of polysulfides in lithium-sulfur(Li-S)batteries are a significant obstacle to their widespread adoption as energy storage devices.However,recent studies have shown that tungsten oxide(WO_(3))can facilitate the conversion kinetics of polysulfides in Li-S batteries.Herein,we fabricated host materials for sulfur using nitrogen-doped carbon nanotubes(N-CNTs)and WO_(3).We used low-cost components and simple procedures to overcome the poor electrical conductivity that is a disadvantage of metal oxides.The composites of WO_(3) and N-CNTs(WO_(3)/N-CNTs)create a stable framework structure,fast ion diffusion channels,and a 3D electron transport network during electrochemical reaction processes.As a result,the WO_(3)/N-CNT-Li2S6 cathode demonstrates high initial capacity(1162 mA·h·g^(-1) at 0.5℃),excellent rate performance(618 mA·h·g^(-1) at 5.5℃),and a low capacity decay rate(0.093%up to 600 cycles at 2℃).This work presents a novel approach for preparing tungsten oxide/carbon composite catalysts that facilitate the redox kinetics of polysulfide conversion.

In silico optimization of actuation performance in dielectric elastomercomposites via integrated finite element modeling and deep learning

Dielectric elastomers(DEs)require balanced electric actuation performance and mechanical integrity under applied voltages.Incorporating high dielectric particles as fillers provides extensive design space to optimize concentration,morphology,and distribution for improved actuation performance and material modulus.This study presents an integrated framework combining finite element modeling(FEM)and deep learning to optimize the microstructure of DE composites.FEM first calculates actuation performance and the effective modulus across varied filler combinations,with these data used to train a convolutional neural network(CNN).Integrating the CNN into a multi-objective genetic algorithm generates designs with enhanced actuation performance and material modulus compared to the conventional optimization approach based on FEM approach within the same time.This framework harnesses artificial intelligence to navigate vast design possibilities,enabling optimized microstructures for high-performance DE composites.

Composite Structural Optimization by Genetic Algorithm and Neural Network Response Surface Modeling

Neural-Network Response Surfaces （NNRS） is applied to replace the actual expensive finite element analysis during the composite structural optimization process. The Orthotropic Experiment Method （OEM） is used to select the most appropriate design samples for network training. The trained response surfaces can either be objective function or constraint conditions. Together with other conven- tional constraints, an optimization model is then set up and can be solved by Genetic Algorithm （GA）. This allows the separation between design analysis modeling and optimization searching. Through an example of a hat-stiffened composite plate design, the weight response surface is constructed to be objective function, and strength and buckling response surfaces as constraints; and all of them are trained through NASTRAN finite element analysis. The results of optimization study illustrate that the cycles of structural analysis ean be remarkably reduced or even eliminated during the optimization, thus greatly raising the efficiency of optimization process. It also observed that NNRS approximation can achieve equal or even better accuracy than conventional functional response surfaces.

Efficient Preconstruction of Three‑Dimensional Graphene Networks for Thermally Conductive Polymer Composites

Electronic devices generate heat during operation and require efficient thermal management to extend the lifetime and prevent performance degradation.Featured by its exceptional thermal conductivity,graphene is an ideal functional filler for fabricating thermally conductive polymer composites to provide efficient thermal management.Extensive studies have been focusing on constructing graphene networks in polymer composites to achieve high thermal conductivities.Compared with conventional composite fabrications by directly mixing graphene with polymers,preconstruction of three-dimensional graphene networks followed by backfilling polymers represents a promising way to produce composites with higher performances,enabling high manufacturing flexibility and controllability.In this review,we first summarize the factors that affect thermal conductivity of graphene composites and strategies for fabricating highly thermally conductive graphene/polymer composites.Subsequently,we give the reasoning behind using preconstructed three-dimensional graphene networks for fabricating thermally conductive polymer composites and highlight their potential applications.Finally,our insight into the existing bottlenecks and opportunities is provided for developing preconstructed porous architectures of graphene and their thermally conductive composites.

Fabrication and abrasive wear properties of metal matrix composites reinforced with three-dimensional network structure

Reticulated polyurethane was chosen as the preceramic material for preparing the porous preform using the replication process. The immersing and sintering processes were each performed twice for fabricating a high-porosity and super-strong skeleton. The aluminum magnesium matrix composites reinforced with three-dimensional network structure were prepared using the infiltration technique by pressure assisting and vacuum driving. Light interfacial reactions have played a profitable role in most of the ceramic-metal systems. The metal matrix composites interpenetrated with the ceramic phase have a higher wear resistance than the metal matrix phase. The volume fraction of ceramic reinforcement has a significant effect on the abrasive wear, and the wear rate can be decreased with the increase of the volume fraction of reinforcement.

Neural Network-Based Second Order Reliability Method(NNBSORM)for Laminated Composite Plates in Free Vibration

Monte Carlo Simulations(MCS),commonly used for reliability analysis,require a large amount of data points to obtain acceptable accuracy,even if the Subset Simulation with Importance Sampling(SS/IS)methods are used.The Second Order Reliability Method(SORM)has proved to be an excellent rapid tool in the stochastic analysis of laminated composite structures,when compared to the slower MCS techniques.However,SORM requires differentiating the performance function with respect to each of the random variables involved in the simulation.The most suitable approach to do this is to use a symbolic solver,which renders the simulations very slow,although still faster than MCS.Moreover,the inability to obtain the derivative of the performance function with respect to some parameters,such as ply thickness,limits the capabilities of the classical SORM.In this work,a Neural Network-Based Second Order Reliability Method(NNBSORM)is developed to replace the finite element algorithm in the stochastic analysis of laminated composite plates in free vibration.Because of the ability to obtain expressions for the first and second derivatives of the NN system outputs with respect to any of its inputs,such as material properties,ply thicknesses and orientation angles,the need for using a symbolic solver to calculate the derivatives of the performance function no longer exists.The proposed approach is accordingly much faster,and easily allows for the consideration of ply thickness uncertainty.The present analysis showed that dealing with ply thicknesses as random variables results in 37%increase in the laminate’s probability of failure.

Regulatable Orthotropic 3D Hybrid Continuous Carbon Networks for Efficient Bi-Directional Thermal Conduction

Vertically oriented carbon structures constructed from low-dimen-sional carbon materials are ideal frameworks for high-performance thermal inter-face materials(TIMs).However,improving the interfacial heat-transfer efficiency of vertically oriented carbon structures is a challenging task.Herein,an orthotropic three-dimensional(3D)hybrid carbon network(VSCG)is fabricated by depositing vertically aligned carbon nanotubes(VACNTs)on the surface of a horizontally oriented graphene film(HOGF).The interfacial interaction between the VACNTs and HOGF is then optimized through an annealing strategy.After regulating the orientation structure of the VACNTs and filling the VSCG with polydimethylsi-loxane(PDMS),VSCG/PDMS composites with excellent 3D thermal conductive properties are obtained.The highest in-plane and through-plane thermal conduc-tivities of the composites are 113.61 and 24.37 W m^(-1)K^(-1),respectively.The high contact area of HOGF and good compressibility of VACNTs imbue the VSCG/PDMS composite with low thermal resistance.In addition,the interfacial heat-transfer efficiency of VSCG/PDMS composite in the TIM performance was improved by 71.3%compared to that of a state-of-the-art thermal pad.This new structural design can potentially realize high-performance TIMs that meet the need for high thermal conductivity and low contact thermal resistance in interfacial heat-transfer processes.

Machine learning-based stiffness optimization of digital composite metamaterials with desired positive or negative Poisson's ratio

Mechanical metamaterials such as auxetic materials have attracted great interest due to their unusual properties that are dictated by their architectures.However,these architected materials usually have low stiffness because of the bending or rotation deformation mechanisms in the microstructures.In this work,a convolutional neural network(CNN)based self-learning multi-objective optimization is performed to design digital composite materials.The CNN models have undergone rigorous training using randomly generated two-phase digital composite materials,along with their corresponding Poisson's ratios and stiffness values.Then the CNN models are used for designing composite material structures with the minimum Poisson's ratio at a given volume fraction constraint.Furthermore,we have designed composite materials with optimized stiffness while exhibiting a desired Poisson's ratio(negative,zero,or positive).The optimized designs have been successfully and efficiently obtained,and their validity has been confirmed through finite element analysis results.This self-learning multi-objective optimization model offers a promising approach for achieving comprehensive multi-objective optimization.

High Temperature Flow Stress Prediction of Nano-Al_2O_3/Cu Composite Using an Artificial Neural Network

Alumina dispersion strengthened copper composite （nano-Al2O3/Cu composite） was recently emerged as a kind of potentially viable and attractive engineering material for applications requiring high strength, high thermal and electrical conductivities and resistance to softening at elevated temperatures. The nano-Al2O3/Cu composite was produced by internal oxidation. The microstructures of the composite were analyzed by the TEM and its hot deformation behavior was investigated by means of continuous compression tests performed on a Gleeble 1500 thermo-simulator. Making use of the modified algorithm-Levenberg-Marquardt （L-M） algorithm BP neural network, a model for predicting the flow stresses during hot deformation was set up on the base of the experimental data. Results show that the microstructures of the composite are characterized by uniform distribution of nano-Al2O3 particles in Cu-matrix. The sliding of dislocations is the main deformation mechanism. The dynamic recovery is the main softening mode with the flow stress decreasing gently from 500℃ to 850 ~C. The recrystallization of Cu-matrix can be retarded late into as high as 850 ℃, when it happens only partially. The well-trained BP neural network model can accurately describe the influence of the temperature, strain rate, and true strain on the flow stresses, therefore, it can precisely predict the flow stresses of the composite under given deforming conditions and provide a new way to optimize hot deforming process parameters.

Application of BP Neural Network in Prediction of Cu-Pb Composite Plates Properties

： The artificial neural networks（ANN） , which have broad application, are proposed to develop Cu-Pb composite plates materials. Based on the back propagation（BP） algorithm of the forward muhilayer perceptron, the model to predict the shear stress under different ingredient of the third element and the hot dipping temperature for Cu-Pb composite plates are established. Then the relational model among the third element, hot dipping temperature and shear stress by using the limited data are studied, and the forecast average error is 4%. This model can satisfy the requirements of the precision of forecast in the project experiment process. The results show that the corresponding shear stress is greater when the third element in the element contains more Sn; the most appropriate temperature of hot-dip plating about is 340℃, ＇after predicted with lead/the third element/the best performance of copper composite material element of the third group is the one-element Sn, hot dip plating temperature is 335 ℃ ; two-element is 90% Sn 10% Bi, and hot dip plating temperature is 345 ℃. The prediction results can be used for a reference in instructing the further experimental design.

A Composite Service Provision Method Based on Novel Node Model in Mobile Ad Hoc Networks

Composite service provision in mobile ad hoc networks encounters great challenges and its success rate is not satisfactory because the nodes' locations are dynamic and the nodes maybe unavailable at any time.Composite service is built through the service composition.In mobile ad hoc networks,the factors influencing the success rate of service composition are mainly the number of nodes and the time spent for the composition.The node's failure probability is proportional to the length of time the node exist in the networks.In order to improve the success rate of service composition,we take several measures.First,we split the service requirement into several segments and cluster the nodes,so that the nodes' waiting time for service composition can be reduced.Second,we propose a new node model of "one node contains multiple services" in mobile ad hoc networks.Using this type of nodes model,the number of nodes required for service composition can be reduced.These means can increase the success rate of service composition.

Automatic well test interpretation based on convolutional neural network for a radial composite reservoir

An automatic well test interpretation method for radial composite reservoirs based on convolutional neural network(CNN) is proposed, and its effectiveness and accuracy are verified by actual field data. In this paper, based on the data transformed by logarithm function and the loss function of mean square error(MSE), the optimal CNN is obtained by reducing the loss function to optimize the network with "dropout" method to avoid over fitting. The trained optimal network can be directly used to interpret the buildup or drawdown pressure data of the well in the radial composite reservoir, that is, the log-log plot of the given measured pressure variation and its derivative data are input into the network, the outputs are corresponding reservoir parameters(mobility ratio, storativity ratio, dimensionless composite radius, and dimensionless group characterizing well storage and skin effects), which realizes the automatic initial fitting of well test interpretation parameters. The method is verified with field measured data of Daqing Oilfield. The research shows that the method has high interpretation accuracy, and it is superior to the analytical method and the least square method.

Design and Construction of Composite Tubular Arches with Network Suspension System： Recent Undertakings and Trends

The use of Network hanger arrangement, a development of the classical Nielsen V-hanger system, in steel bowstring arch bridges allows for important steel saving, with very slender main elements, owing to the remarkable reduction of bending stresses in the arches and tie beams. The present paper describes the main features of the design and construction of several long-span arch bridges of this typology in Spain： the three pedestrian footbridges for the Madrid cycling ring track, with spans of 52, 60 and 80 m, the Bridge over River Deba in Guipuzcoa with a span of 110 m and Palma del Rio Bridge over River Guadalquivir in Cordoba, 130 m long. In all cases, two inclined arches linked at the crown were implemented, a very effective disposition to reduce the out-of-plane buckling length. The multiple crossings of the hanger system, consisting of prestressed bars in the case of Deba Bridge and the footbridges, and locked coil cables for Palma del Rio Bridge, were dealt with by means of crossing devices which led to a technically satisfactory solution with minimal visual impact. An innovative approach to bowstring arches was introduced in Valdebebas Bridge over M-12 motorway in Madrid, next to the new T-4 Terminal of Barajas Airport, with a span of 162 m, where the hangers are replaced by a structural steel mesh -diagrid- which acts as the web of a simply-supported beam whose compression head is the arch and the tie beam is the deck.

Preparation and electrochemical properties of LiFePO_4/C composite with network structure for lithium ion batteries

The bare LiFePO4 and LiFePO4/C composites with network structure were prepared by solid-state reaction. The crystalline structures, morphologies and specific surface areas of the materials were investigated by X-ray diffractometry(XRD), scanning electron microscopy(SEM) and multi-point brunauer emmett and teller(BET) method. The results show that the LiFePO4/C composite with the best network structure is obtained by adding 10% phenolic resin carbon. Its electronic conductivity increases to 2.86×10-2 S/cm. It possesses the highest specific surface area of 115.65 m2/g, which exhibits the highest discharge specific capacity of 164.33 mA·h/g at C/10 rate and 149.12 mA·h/g at 1 C rate. The discharge capacity is completely recovered when C/10 rate is applied again.

A step to the decentralized real-time timekeeping network

The composite time scale(CTS) provides an accurate and stable time-frequency reference for modern science and technology. Conventional CTS always features a centralized network topology, which means that the CTS is accompanied by a local master clock. This largely restricts the stability and reliability of the CTS. We simulate the restriction and analyze the influence of the master clock on the CTS. It proves that the CTS's long-term stability is also positively related to that of the master clock, until the region dominated by the frequency drift of the H-maser(averaging time longer than ~10~5s).Aiming at this restriction, a real-time clock network is utilized. Based on the network, a real-time CTS referenced by a stable remote master clock is achieved. The experiment comparing two real-time CTSs referenced by a local and a remote master clock respectively reveals that under open-loop steering, the stability of the CTS is improved by referencing to a remote and more stable master clock instead of a local and less stable master clock. In this way, with the help of the proposed scheme, the CTS can be referenced to the most stable master clock within the network in real time, no matter whether it is local or remote, making democratic polycentric timekeeping possible.

Chien-physics-informed neural networks for solving singularly perturbed boundary-layer problems

A physics-informed neural network(PINN)is a powerful tool for solving differential equations in solid and fluid mechanics.However,it suffers from singularly perturbed boundary-layer problems in which there exist sharp changes caused by a small perturbation parameter multiplying the highest-order derivatives.In this paper,we introduce Chien's composite expansion method into PINNs,and propose a novel architecture for the PINNs,namely,the Chien-PINN(C-PINN)method.This novel PINN method is validated by singularly perturbed differential equations,and successfully solves the wellknown thin plate bending problems.In particular,no cumbersome matching conditions are needed for the C-PINN method,compared with the previous studies based on matched asymptotic expansions.

In situ(α-Al_2O_3+ZrB_2)/Al composites with network distribution fabricated by reaction hot pressing

In situ（α-Al2O3+ZrB2）/Al composites with network distribution were fabricated using low-energy ball milling and reaction hot pressing. Differential thermal analysis（DTA） was used to study the reaction mechanisms in the Al–Zr O2–B system. X-ray diffraction（XRD） and scanning electron microscopy（SEM） in conjunction with energy-dispersive X-ray spectroscopy（EDX） were used to investigate the composite phases, morphology, and microstructure of the composites. The effect of matrix network size on the microstructure and mechanical properties was investigated. The results show that the optimum sintering parameters to complete reactions in the Al–Zr O2–B system are 850°C and 60 min. In situ-synthesized α-Al2O3 and Zr B2 particles are dispersed uniformly around Al particles, forming a network microstructure; the diameters of the α-Al2O3 and Zr B2 particles are approximately 1–3 μm. When the size of Al powder increases from 60–110 μm to 150–300 μm, the overall surface contact between Al powders and reactants decreases, thereby increasing the local volume fraction of reinforcements from 12% to 21%. This increase of the local volume leads to a significant increase in microhardness of the in situ（α-Al2O3–Zr B2）/Al composites from Hv 163 to Hv 251.