Knowledge-reused transfer learning for molecular and materials science

Leveraging big data analytics and advanced algorithms to accelerate and optimize the process of molecular and materials design, synthesis, and application has revolutionized the field of molecular and materials science, allowing researchers to gain a deeper understanding of material properties and behaviors,leading to the development of new materials that are more efficient and reliable. However, the difficulty in constructing large-scale datasets of new molecules/materials due to the high cost of data acquisition and annotation limits the development of conventional machine learning(ML) approaches. Knowledgereused transfer learning(TL) methods are expected to break this dilemma. The application of TL lowers the data requirements for model training, which makes TL stand out in researches addressing data quality issues. In this review, we summarize recent progress in TL related to molecular and materials. We focus on the application of TL methods for the discovery of advanced molecules/materials, particularly, the construction of TL frameworks for different systems, and how TL can enhance the performance of models. In addition, the challenges of TL are also discussed.

A transfer learning enhanced physics-informed neural network for parameter identification in soft materials

Soft materials,with the sensitivity to various external stimuli,exhibit high flexibility and stretchability.Accurate prediction of their mechanical behaviors requires advanced hyperelastic constitutive models incorporating multiple parameters.However,identifying multiple parameters under complex deformations remains a challenge,especially with limited observed data.In this study,we develop a physics-informed neural network(PINN)framework to identify material parameters and predict mechanical fields,focusing on compressible Neo-Hookean materials and hydrogels.To improve accuracy,we utilize scaling techniques to normalize network outputs and material parameters.This framework effectively solves forward and inverse problems,extrapolating continuous mechanical fields from sparse boundary data and identifying unknown mechanical properties.We explore different approaches for imposing boundary conditions(BCs)to assess their impacts on accuracy.To enhance efficiency and generalization,we propose a transfer learning enhanced PINN(TL-PINN),allowing pre-trained networks to quickly adapt to new scenarios.The TL-PINN significantly reduces computational costs while maintaining accuracy.This work holds promise in addressing practical challenges in soft material science,and provides insights into soft material mechanics with state-of-the-art experimental methods.

Transfer matrix method for free and forced vibrations of multi-level functionally graded material stepped beams with different boundary conditions

Functionally graded materials(FGMs)are a novel class of composite materials that have attracted significant attention in the field of engineering due to their unique mechanical properties.This study aims to explore the dynamic behaviors of an FGM stepped beam with different boundary conditions based on an efficient solving method.Under the assumptions of the Euler-Bernoulli beam theory,the governing differential equations of an individual FGM beam are derived with Hamilton’s principle and decoupled via the separation-of-variable approach.Then,the free and forced vibrations of the FGM stepped beam are solved with the transfer matrix method(TMM).Two models,i.e.,a three-level FGM stepped beam and a five-level FGM stepped beam,are considered,and their natural frequencies and mode shapes are presented.To demonstrate the validity of the method in this paper,the simulation results by ABAQUS are also given.On this basis,the detailed parametric analyses on the frequencies and dynamic responses of the three-level FGM stepped beam are carried out.The results show the accuracy and efficiency of the TMM.

Numerical Modelling of Coupled Heat and Mass Transfer in Porous Materials: Application to Cinder Block Bricks

In this work, we present numerical modelling of coupled heat and mass transfer within porous materials. Our study focuses on cinder block bricks generally used in building construction. The material is assumed to be placed in air. Moisture content and temperature have been chosen as the main transfer drivers and the equations governing these transfer drivers are based on the Luikov model. These equations are solved by an implicit finite difference scheme. A Fortran code associated with the Thomas algorithm was used to solve the equations. The results show that heat and mass transfer depend on the temperature of the air in contact with the material. As this air temperature rises, the temperature within the material increases, and more rapidly at the material surface. Also, thermal conductivity plays a very important role in the thermal conduction of building materials and influences heat and mass transfer in these materials. Materials with higher thermal conductivity diffuse more heat.

Improvement of High-Speed Detection Algorithm for Nonwoven Material Defects Based on Machine Vision

Defect detection is vital in the nonwoven material industry,ensuring surface quality before producing finished products.Recently,deep learning and computer vision advancements have revolutionized defect detection,making it a widely adopted approach in various industrial fields.This paper mainly studied the defect detection method for nonwoven materials based on the improved Nano Det-Plus model.Using the constructed samples of defects in nonwoven materials as the research objects,transfer learning experiments were conducted based on the Nano DetPlus object detection framework.Within this framework,the Backbone,path aggregation feature pyramid network(PAFPN)and Head network models were compared and trained through a process of freezing,with the ultimate aim of bolstering the model's feature extraction abilities and elevating detection accuracy.The half-precision quantization method was used to optimize the model after transfer learning experiments,reducing model weights and computational complexity to improve the detection speed.Performance comparisons were conducted between the improved model and the original Nano Det-Plus model,YOLO,SSD and other common industrial defect detection algorithms,validating that the improved methods based on transfer learning and semi-precision quantization enabled the model to meet the practical requirements of industrial production.

Heat transfer and parametric studies of an encapsulated phase change material based cool thermal energy storage system

This work investigates the transient behaviour of a phase change material based cool thermal energy storage (CTES) system comprised of a cylindrical storage tank filled with encapsulated phase change materials (PCMs) in spherical container integrated with an ethylene glycol chiller plant. A simulation program was developed to evaluate the temperature histories of the heat transfer fluid (HTF) and the phase change material at any axial location during the charging period. The results of the model were validated by comparison with experimental results of temperature profiles of HTF and PCM. The model was also used to investigate the effect of porosity, Stanton number, Stefan number and Peclet number on CTES system performance. The results showed that increase in porosity contributes to a higher rate of energy storage. However, for a given geometry and heat transfer coefficient, the mass of PCM charged in the unit decreases as the increase in porosity. The St number as well as the Ste number is also influential in the performance of the unit. The model is a convenient and more suitable method to determine the heat transfer characteristics of CTES system. The results reported are much useful for designing CTES system.

Material transfer behavior of AgTiB_2 contact under different contact forces and electrode gaps

To disclose the effect of contact force and electrode gap on the material transfer behavior of Ag-based contact material, arc-erosion tests of the Ag-4wt.%TiB2 contact material were performed for 5000 operations at 24 V/16 A under resistive load on an electric contact material testing system. The arc energy and arc duration were investigated, the surface morphologies of eroded anode and cathode were characterized, the mass changes after arc-erosion tests were determined, and the material transfer behavior was discussed as well. The results show that contact force has a significant effect on the arc energy, arc duration and erosion morphology, but has no impact on the material transfer mode. However, electrode gap not only influences the arc energy, arc duration and surface morphology, but also changes the material transfer mode. At 1 mm, the material transfers from anode to cathode. Nevertheless, an opposite mode presents at 4 mm, which is from cathode to anode.

Study of the Material Transfer Characteristics and Surface Morphology Due to Arc Erosion of PtIr Contact Materials

By means of breaking tests on PtIr contact materials via a JF04C contact material testing machine, it was attempted to elucidate the characteristics of the various surface morphology and material transfer after the arc erosion process caused by break arc. The material transfer characteristics appeared in the experiments were concluded and analyzed. Meanwhile, the morphology of the anode and cathode surface were observed and analyzed by SEM.

Near-field radiative heat transfer in hyperbolic materials

In the post-Moore era, as the energy consumption of micro-nano electronic devices rapidly increases, near-field radiative heat transfer(NFRHT) with super-Planckian phenomena has gradually shown great potential for applications in efficient and ultrafast thermal modulation and energy conversion. Recently, hyperbolic materials, an important class of anisotropic materials with hyperbolic isofrequency contours, have been intensively investigated. As an exotic optical platform, hyperbolic materials bring tremendous new opportunities for NFRHT from theoretical advances to experimental designs. To date, there have been considerable achievements in NFRHT for hyperbolic materials, which range from the establishment of different unprecedented heat transport phenomena to various potential applications. This review concisely introduces the basic physics of NFRHT for hyperbolic materials, lays out the theoretical methods to address NFRHT for hyperbolic materials, and highlights unique behaviors as realized in different hyperbolic materials and the resulting applications. Finally, key challenges and opportunities of the NFRHT for hyperbolic materials in terms of fundamental physics, experimental validations, and potential applications are outlined and discussed.

Analysis of material flow and heat transfer in friction stir welding of aluminium alloys

A three-dimensional viscous-plastic.finite element model is established based on computational fluid mechanics. The material during the welding process is considered as non-Newtonian fluid abided by Norton-Hoff constitutive law, and viscous dissipation is assumed as the unique heat source. The model is used to numerically simulate the material flow and heat transfer in friction stir welding, and capture some useful process characteristics, .such as heat generation, temperature distribution and fluid.flow; besides, the velocity field is used to calculate streamlines of material flow, and the dimension of the deformation zone is measured.

Effect of different heat transfer fluids on discharging performance of phase change material included cylindrical container during forced convection

In the present work,effects of various heat transfer fluids on the discharging performance of a phase change material(PCM) included cylindrical container are numerically assessed during forced convection.The heat transfer fluid air,hydrogen,water and nanofluid with alumina particles are used and the the geometric variation of the PCM embedded region is also considered.The finite element method is used as the solver.Dynamic features of heat exchange with various phases are explored for different heat transfer fluid types,Reynolds number(between 100 and 300) and PCM embedded region geometric variation(h_(x)between 0.01 d_(1) and 0.65 d_(1),hybetween 0.1 h_(1) and 0.4 h_(1)).It is observed that discharging time is significantly influenced by the heat transfer fluid type while full phase transition time for air is obtained as more than 10 times when hydrogen is utilized as heat transfer fluid.The best performance is achieved with nanofluid.When the PCM integrated region size is reduced,discharging time is generally reduced while due to the form of the geometry,vortex formation is established in the PCM region.This results in performance degeneration at the highest radius and height of the inner cylinder.Discharging time increases by about 12% when radius of the inner cylinder is increased from h_(x)=0.35 d_(1) to h_(x)=0.45 d_(1).Dynamic features of PCM temperature and liquid fraction are affected with Reynolds number while discharging time is reduced by about 48% when configurations with the lowest and highest Reynolds number are compared.

Supplementary Materials: Ultrafast charge transfer in dual graphene-WS_2 van der Waals quadrilayer heterostructures

1. The transient absorption spectra of the WS2 monolayer sample.In the measurement of the transient absorption spectra of the WS2 monolayer sample, A 400-nm (3.1eV) pump pulse with a peak fluence of about 10μJ/cm2excites the electrons from the valence band into the conduction band,the

Concerning The Effect of Surface Material on Nucleate Boiling Heat Transfer of R-113

This paper presents results of an experimental investigation carried out to determine the effects of surface material on nucleate pool boiling heat transfer of refrigerant R113. Experiments were performed on horizontal circular plates of brass, copper and aluminum. The heat transfer coefficient was evaluated by measuring wall superheat and effective heat flux removed by boiling. The experiments were carried out in the heat flux range of 8 to 200 kW/m2. The obtained results have shown significant effect of surface material, with copper providing the highest heat transfer coefficient among the samples, and aluminum the least. There was negligible difference at low heat fluxes, but copper showed 23% better performance at high heat fluxes than aluminum and 18% better than brass.

Heat Transfer Characteristics of Work Fluid Including Phase Change Material That Flow into Heating Surface from Narrow Path

Use of the low temperature (less than 100°C) energy contributes to effective use of heat resources. The cost recovery by power generation is difficult by using an existing system (the binary cycle or the thermoelectric conversion element), because the initial investment is large. The final purpose of this research is development of the low temperature difference drive engine supposing use in a hot-springs resort as a power source for electric power generation. In order that a traveler may look at and delight a motion of an engine, it is made to drive at low-speed number of rotations. An engine cycle of this study is aimed at the development of Stirling cycle engine which can maintain high efficiency in small size. This kind of engine has simple structure;it brings low cost, and it is easy to perform maintenance. However, it is difficult to obtain enough output by this type of engine, because of its low temperature difference. This paper deals with the heat transfer characteristic that the working fluid including a phase change material flows into the heating surface from the narrow path. In order to increase the amount of the heat transmission, Diethylether is added to the working fluid. Diethylether is selected as a phase change material (PCM) that has the boiling point which exists between the heat source of high temperature and low temperature. The parameters of the experiment are additive amount of PCM, rotational speed of the displacer piston and temperature of heat transfer surface. It is shown that it is possible to make exchange of heat amount increase by adding phase change material. The result of this research shows the optimal condition of the difference in temperature in heat processing, number of revolutions, and addition concentration of PCM.

Second-order two-scale computations for conductive radiative heat transfer problem in periodic porous materials

In this paper, a kind of second-order two-scale （SOTS） computation is developed for conductive-radiative heat trans- fer problem in periodic porous materials. First of all, by the asymptotic expansion of the temperature field, the cell problem, homogenization problem, and second-order correctors are obtained successively. Then, the corresponding finite element al- gorithms are proposed. Finally, some numerical results are presented and compared with theoretical results. The numerical results of the proposed algorithm conform with those of the FE algorithm well, demonstrating the accuracy of the present method and its potential applications in thermal engineering of porous materials.

Unsteady Heat Transfer in Bilayer,and Three-Layer Materials

The heat transfer equation is used to determine the heat flow by conduction through a composite material along the real axis.An analytical dimensionless analysis is implemented in the framework of a separation of variables method(SVM).This approach leads to an Eigenvalues problem that is solved by the Newton’s method.Two types of dynamics are found:An unsteady condition(in the form of jumps or drops in temperatures depending on the considered case),and a permanent equilibrium(tending to the ambient temperature).The validity and effectiveness of the proposed approach for any number of adjacent layers is also discussed.It is shown that,as expected,the diffusion of the temperature is linked to the ratio of the thermo-physical properties of the considered layers and their number.

Finite element formulation of axisymmetric heat transfer problem for orthotropic materials

By using Galerkin’s method, the finite element formulation is made for axisymmtric heat transfer problems for anisotropic materials from the heat transfer differential equations expressed in terms of heat fluid density. Results of an example show that the heat transfer anisotropy has an important effect on temperature field.

Time Dependent Surface Heat Transfer in Light Weight Aggregate Cement Based Materials

Surface temperature changes of building materials affect the calculation of heat flow and thus the energy use in heating and cooling. The surface heat transfer coefficient , limiting the heat flow between material surface and ambient air is normally taken as a constant. In this study we propose a time-dependent function . We estimate from unidirectional heat flow experiments with transient and steady-state conditions. Using temperature measurements and the conservation of energy at the surface including convective and irradiative boundary conditions, the value of was obtained both using Finite Difference and Taylor Polynomials methods. Numerical solutions of temperature distribution as function of time were improved with the obtained -functions compared to with constant . There were no clear difference between on different materials, and the final values observed were in the order of magnitude expected from the literature.

Design of transfer device for magnetic thin sheet material

In order to realize the processing and retrieval of magnetic thin sheet materials in industrial production, this paper proposed a kind of transfer device for magnetic thin sheet raw material, which uses rodless cylinders as the main motive device and the programmable logic controller （PLC） to achieve the required functions, and applies the finite element analysis method to analyze its main components in the end of the design.

Visualization and simulation of plastic material flow in friction stir welding of 2024 aluminium alloy plates

Thin copper sheets as marker material were embedded into weld path of 2024 aluminium alloy plates and their final position after friction stir welding was examined by metallographic techniques. Referring to the visualized material flow patterns, a three-dimensional model was developed to conduct the numerical simulation of the temperature profile and plastic material flow in friction stir welding. The calculated velocity contour of plastic flow in close proximity of the tool is generally consistent with the visualized results. As the tool rotation speed increases at a constant tool travel speed, the material flow near the pin gets stronger. The predicted shape and size of the weld nugget zone match with the experimentally measured ones.