Overview:recent studies of machine learning in phase prediction of high entropy alloys

High entropy alloys(HEAs),especially refractory HEAs,have become a subject of interest in the past years due to their exceptional properties in terms of high-temperature strength,corrosion resistance,radiation tolerance,etc.under extreme environments.While the phase formation of these alloys significantly affects their properties.If the phase of HEAs can be forecasted before the experiments,the material design process can be greatly accelerated.The phase formation study of HEAs mainly relied on trial-and-error experiments and multi-scale computational simulations such as calculation of phase diagrams(CALPHAD) and density functional theory(DFT).However,those methods require massive time,man-power,and resources.As a highly efficient tool,machine learning(ML) method has been developed and applied to predict the phase formation of HEAs very recently.This review provided a comprehensive overview and analysis of the most recent research work in this area.First,we introduce ML methodologies applied in HEAs’ phase prediction in terms of principles,database,algorithm,and validation.We then summarize recent applications of the ML method in the phase prediction of HEAs.In the end,we propose possible solutions to the current problems and future research pathways for various challenges in the phase prediction of HEAs using ML.

Phase prediction for high-entropy alloys using generative adversarial network and active learning based on small datasets

In this paper,a new machine learning(ML)model combining conditional generative adversarial networks(CGANs)and active learning(AL)is proposed to predict the body-centered cubic(BCC)phase,face-centered cubic(FCC)phase,and BCC+FCC phase of high-entropy alloys(HEAs).Considering the lack of data,CGANs are introduced for data augmentation,and AL can achieve high prediction accuracy under a small sample size owing to its special sample selection strategy.Therefore,we propose an ML framework combining CGAN and AL to predict the phase of HEAs.The arithmetic optimization algorithm(AOA)is introduced to improve the artificial neural network(ANN).AOA can overcome the problem of falling into the locally optimal solution for the ANN and reduce the number of training iterations.The AOA-optimized ANN model trained by the AL sample selection strategy achieved high prediction accuracy on the test set.To improve the performance and interpretability of the model,domain knowledge is incorporated into the feature selection.Additionally,considering that the proposed method can alleviate the problem caused by the shortage of experimental data,it can be applied to predictions based on small datasets in other fields.

Numerical modeling of global seismic phases and its application in seismic phase identification

Earthquake data include informative seismic phases that require identification for imaging the Earth's structural interior.In order to identify the phases,we created a numerical method to calculate the traveltimes and raypaths by a shooting technique based upon the IASP91 Earth model,and it can calculate the traveltimes and raypaths for not only the seismic phases in the traditional traveltime tables such as IASP91,AK135,but also some phases such as pPcP,pPKIKP,and PPPPP.It is not necessary for this method to mesh the Earth model,and the results from the numerical modeling and its application show that the absolute differences between the calculated and theoretical traveltimes from the ISAP91 tables are less than 0.1 s.Thus,it is simple in manipulation and fast in computation,and can provide a reliable theoretical prediction for the identification of a seismic phase within the acquired earthquake data.

Multi-Level Inverter Linear Predictive Phase Composition Strategy for UPQC

The power system is facing numerous issues when the distributed gen-eration is added to the existing system.The existing power system has not been planned with flawless power quality control.These restrictions in the power trans-mission generation system are compensated by the use of devices such as the Static Synchronous Compensator(STATCOM),the Unified Power Quality Con-ditioner(UPQC)series/shunt compensators,etc.In this work,UPQC’s plan with the joint activity of photovoltaic(PV)exhibits is proposed.The proposed system is made out of series and shunt regulators and PV.A boost converter connects the DC link to the PV source,allowing it to compensate for voltage sags,swells,vol-tage interferences,harmonics,and reactive power issues.In this paper,the fea-tures of a seven-level Cascaded H-Bridge Multi-Level idea are applied to shunt and series active filter changeovers to reduce Total Harmonic Distortion and com-pensate for voltage issues.Despite its power quality capacity for common cou-pling,the proposed system can inject the grid’s dynamic power.During voltage interference,it can also provide a piece of delicate burden power.The simulation is carried out with the help of MATLAB/SIMULINK programming,and the results are compared to those of other conventional methods.

A Mathematical Model of Heat Transfer in Problems of Pipeline Plugging Agent Freezing Induced by Liquid Nitrogen

A mathematical model for one-dimensional heat transfer in pipelines undergoing freezing induced by liquid nitrogen is elaborated.The basic premise of this technology is that the content within a pipeline is frozen to form a plug or two plugs at a position upstream and downstream from a location where work a modification or a repair must be executed.Based on the variable separation method,the present model aims to solve the related coupled heat conduction and moving-boundary phase change problem.An experiment with a 219 mm long pipe,where water was taken as the plugging agent,is presented to demonstrate the relevance and reliability of the proposed model(results show that the error is within 18%).Thereafter,the model is applied to predict the cooling and freezing process of pipelines with different inner diameters at different liquid nitrogen refrigeration temperatures when water is used as the plugging agent.

High-pressure structures of InBi predicted by particle swarm optimization algorithm

We extensively explore the high-pressure structures of InBi by using a newly developed particle swarm optimization algorithm. An orthorhombic Imma structure is discovered to be stable from 43.7 GPa to 107.9 GPa, ruling out the previously speculated cubic structure. Further increasing the pressure, we find a tetragonal P4/nmm structure which is energetically more favourable from 107.9 CPa to 200 GPa. Especially, the tetragonal P4/nmm structure is known to occur at high pressure in the structures of ZnO and MgTe. We also predict this structure to be a high-pressure structure of ZnTe. Thus the tetragonal P4/nmm structure may be a universal high-pressure structure of the Ⅱ-Ⅵ and the Ⅲ-Ⅴ compounds.

Intelligent Space All-Optical Network Technology

Microwave transmission in a space network is greatly restricted due to precious radio spectrum resources. To meet the demand for large-bandwidth, global seamless coverage and on-demanding access, the Space All-Optical Network(SAON) becomes a promising paradigm. In this paper, the related space optical communications and network programs around the world are first briefly introduced. Then the intelligent Space All-Optical Network(i-SAON), which can be deemed as an advanced SAON, is illustrated, with the emphasis on its features of high survivability, sensing and reconfiguration intelligence, and large capacity for all optical load and switching. Moreover, some key technologies for i-SAON are described, including the rapid adjustment and control of the laser beam direction, the deep learning-based multi-path anti-fault routing, the intelligent multi-fault diagnosis and switching selection mechanism, and the artificial intelligence-based spectrum sensing and situational forecasting.

Dual phase titanium alloy hot forging process design： experiments and numerical modeling

Titanium alloys materials when both good weight reduction are required are considered desirable mechanical properties and at the same time. This class of materials is widely used in those fields （aeronautics, aerospace） in which common steels and light-weight materials, e.g., aluminum alloys, are not able to satisfy all operative service conditions. During the last decade, forg- ing of titanium alloys has attracted greater attention from both industrial and scientific/academic researchers because of their potential in providing a near net shaped part with minimal need for machining. In this paper, a numerical model of the forging sequences for a Ti-6A1-4V titanium alloy aerospace component is presented. The model was tested and validated against experimental forgings. The model is then applied to predict loads final microstructure and defects of an aeronautical component. In addition to metal flow and die stresses, microstructural transformations （α and β phases） are considered for the determination of proper process parameters. It is found that transformation from α/β to β phase during forging and reverse transformations in post-forge cooling needs to be considered in the computational model for reasonable prediction of forging loads and product properties.

A MODEL FOR PREDICTING PHASE INVERSION IN OIL-WATER TWO-PHASE PIPE FLOW

Experiments of phase inversion characteristics for horizontal oil-water two-phase flow in a stainless steel pipe loop （25.7 mm inner diameter,52 m long） are conducted. A new viewpoint is brought forward about the process of phase inversion in oil-water two-phase pipe flow. Using the relations between the total free energies of the pre-inversion and post-inversion dispersions, a model for predicting phase inversion in oil-water two-phase pipe flow has been developed that considers the characteristics of pipe flow. This model is compared against other models with relevant data of phase inversion in oil-water two-phase pipe flow. Results indicate that this model is better than other models in terms of calculation precision and applicability. The model is useful for guiding the design for optimal performance and safety in the operation of oil-water two-phase pipe flow in oil fields.

Additive manufacturing of high-entropy alloys by thermophysical calculations and in situ alloying

Electron beam melting(EBM)is a promising technology to manufacture various alloys with outstanding properties;however,the number of available alloys is limited.We propose in situ alloying to accelerate the development of advanced and novel alloys,based on thermophysical calculations and CALPHAD approach,during the EBM process.We demonstrate our concept through the design and fabrication of high entropy alloys(HEAs).Three CoCrFeNiMn-xTi(x=0.18,0.50,2.00,in molar%)HEAs are manufactured.EBM-built HEAs achieve a homogeneous distribution of elements while forming multiphase alloys resulted from the hot powder bed.The topological structures formed by secondary phases contribute to an increase in the hardness of EBM-built HEAs up to 900 HV1.Considering alloy design,a systematic analysis on Co Cr Fe Ni Mn-0.18 Ti HEA elucidates the microstructural evolution in detail.These findings provide a deep understanding of in situ alloying and pave the way to develop new alloys specific to the EBM process.

A Theoretical Interpretation of Free Volume at Glass Transition

We device a relaxed lattice model （RLM） to study the mechanism of glass transition, which unifies the cage- effects from particle-particle interaction and entropy. By analyzing entropy in RLM with considering the influence of interactions on equilibrium, we demonstrate that glass transition is a second-order phase transition. For a perfect one- dimensional linked particle system like linear polymer under normal pressure, the free volume at glass transition is rigorously deduced out to be 2.6%, which provides a theoretical basis for the iso-free volume of 2.5% given by Willian, Landel and Ferry （WLF） equation. Extending to system with dead particles linked with higher dimensions like branched or cross-linked chains under positive or negative pressure, free volume at glass transition is varied, based on which we construct a phase diagram of glass transition in the space of free volume-dead particle-pressure. This demonstrates that free volume is not the single parameter determining glass transition, while either dead particles like cross-linked points or external force fields like pressure can vary free volume at the glass transition.