Investigating variability of fatigue indicator parameters of two-phase nickel-based superalloy microstructures

Variability of fatigue properties of Nickel-based superalloys induced by microstructure feature uncertainties is investigated. The microstructure at one material point is described by its grain size and orientation features, as well as the volume fraction of the γ′ phase. Principal component analysis (PCA) is introduced to reduce the dimensionality of the microstructure feature space. PCA and kernel PCA (KPCA) techniques are presented and compared. Reduced representations of input features are mapped to uniform or standard Gaussian distributions through polynomial chaos expansion (PCE) so that the sampling of new microstructure realizations becomes feasible. A crystal plasticity constitutive model is adopted to evaluate fatigue properties of two-phase superalloy microstructures under cyclic loading. The fatigue properties are measured by strain-based fatigue indicator parameters (FIP). Adaptive sparse grid collocation (ASGC) and Monte Carlo (MC) methods are used to establish the relation between microstructure feature uncertainties and the variability of macroscopic properties. Convergence with increasing dimensionality of the reduced surrogate stochastic space is studied. Distributions of FIPs and the convex hulls describing the envelope of these parameters in the presence of microstructure uncertainties are shown.     

Residual stresses of types II and III and their estimation

Residual stresses (RS) are induced in metallic materials by a variety of working and fabrication processes. They are generally classified into three categories (types I, II and III) depending upon their range of influence. Separation of type I and II stresses often requires stress measurements on thin specimens. Type III stresses are related closely to grain fragmentation and micro-strains associated with plastic deformation. Perhaps the oldest and the most rigorous method of estimating these stresses is still the Warren-Averbach analysis developed during the late forties/early fifties. Other techniques involving integral breadths and variance of the profiles have also been developed. In all these methods developed during the early stages, prior to seventies, a major requirement was well-separated non-overlapping profiles. The late sixties and early seventies saw a dramatic increase in computational capabilities with the advent of powerful electronic computers. This led to the introduction of curve-fitting procedures into the field of X-ray diffraction. The most remarkable achievement of this period is the development of the Rietveld Method. Although this method was initially developed to tackle the neutron diffraction profiles, which are as a rule nearly symmetric and Gaussian in nature, the method saw rapid developments during the eighties. At present, techniques based on concepts developed by Rietveld could be applied to essentially asymmetric and non-Gaussian multiple spectral components of X-ray diffraction profiles. Pattern decomposition techniques which separate composite powder diffraction profiles into individual profiles are now available. In combination with single line profile analysis techniques, this provides a powerful tool in the hands of researchers. A typical example of such a single line profile analysis is given.     

Microstructure-property relationships of GRP

Fibre-reinforced composite materials are being used increasingly in critical applications, when the primary function of the material is to support loads, but very high safety margins are commonly used for such applications. Such large safety margins arise from the uncertainties regarding the mechanical behaviour of composite materials. The authors believe that the lack of microstructural definition of composite materials may make a substantial contribution to these uncertainties. In this initial study, relationships are sought between microstructure and properties of a model microstructure. The methods used are applicable to a very wide range of composite materials.     

Numerical study of flow field in new design cyclones with different wall temperature profiles: Comparison with conventional ones

In this paper, numerical study of flow field in the new design cyclones with five different wall temperature profiles are investigated. The new design cyclone is based on the idea of improving cyclone collection efficiency and pressure drop by increasing the vortex length. In this paper, the five wall temperature profiles are as follows: (A) cooling with uniform distribution, (B) without temperature change, (C) heating with uniform distribution, (D) incremental linear heating, (E) reduction linear heating. Results are compared in new design and conventional cyclones. The Reynolds averaged Navier–Stokes equations with Reynolds stress turbulence model (RSM) are solved. The Eulerian-Lagrangian computational procedure is used to predict particles tracking in the cyclones. The velocity fluctuations are simulated using the Discrete Random Walk (DRW).Results show that generally, heating the bottom zone of the cyclones can improve the collection efficiency and reduce the pressure drop while heating the top zone of the cyclones marginally affects the flow field. Moreover, cooling the cyclones reduces the efficiency and causes a higher pressure drop. Among five different wall temperature profiles, C and E profiles can increase the efficiency about 8% and profile C reduces the pressure drop by about 9%. The mentioned values in different conditions including particle diameter, flow rate, etc. can be different.     

X-Ray Diffraction Line Broadening: Modeling and Applications to High-Tc Superconductors

A method to analyze powder-diffraction line broadening is proposed and applied to some novel high-T_c superconductors. Assuming that both size-broadened and strain-broadened profiles of the pure-specimen profile are described with a Voigt function, it is shown that the analysis of Fourier coefficients leads to the Warren-Averbach method of separation of size and strain contributions. The analysis of size coefficients shows that the “hook” effect occurs when the Cauchy content of the size-broadened profile is underestimated. The ratio of volume-weighted and surface-weighted domain sizes can change from ~1.31 for the minimum allowed Cauchy content to 2 when the size-broadened profile is given solely by a Cauchy function. If the distortion co-efficient is approximated by a harmonic term, mean-square strains decrease linearly with the increase of the averaging distance. The local strain is finite only in the case of pure-Gauss strain broadening because strains are then independent of averaging distance. Errors of root-mean-square strains as well as domain sizes were evaluated. The method was applied to two cubic structures with average volume-weighted domain sizes up to 3600 Å, as well as to tetragonal and orthorhombic (La-Sr)₂CuO₄, which exhibit weak line broadenings and highly overlapping reflections. Comparison with the integral-breadth methods is given. Reliability of the method is discussed in the case of a cluster of the overlapping peaks. The analysis of La₂CuO₄ and La_1.85M_0.15CuO₄(M = Ca, Ba, Sr) high-T_c superconductors showed that microstrains and incoherently diffracting domain sizes are highly anisotropic. In the superconductors, stacking-fault probability increases with increasing T_c; microstrain decreases. In La₂CuO₄, different broadening of (h00) and (0k0) reflections is not caused by stacking faults; it might arise from lower crystallographic symmetiy. The analysis of Bi-Cu-O superconductors showed much higher strains in the [001] direction than in the basal a-b plane. This may be caused by stacking disorder along the c-axis, because of the two-dimensional weakly bonded BiO double layers. Results for the specimen containing two related high-T_c phases indicate a possible mechanism for the phase transformation by the growth of faulted regions of the major phase.     

Discovery of marageing steels:machine learning vs.physical metallurgical modelling

Physical metallurgical (PM) and data-driven approaches can be independently applied to alloy design.Steel technology is a field of physical metallurgy around which some of the most comprehensive under-standing has been developed,with vast models on the relationship between composition,processing,microstructure and properties.They have been applied to the design of new steel alloys in the pursuit of grades of improved properties.With the advent of rapid computing and low-cost data storage,a wealth of data has become available to a suite of modelling techniques referred to as machine learning (ML).ML is being emergingly applied in materials discovery while it requires data mining with its adoption being limited by insufficient high-quality datasets,often leading to unrealistic materials design predictions outside the boundaries of the intended properties.It is therefore required to appraise the strength and weaknesses of PM and ML approach,to assess the real design power of each towards designing novel steel grades.This work incorporates models and datasets from well-established literature on marageing steels.Combining genetic algorithm (GA) with PM models to optimise the parameters adopted for each dataset to maximise the prediction accuracy of PM models,and the results were compared with ML models.The results indicate that PM approaches provide a clearer picture of the overall composition-microstructure-properties relationship but are highly sensitive to the alloy system and hence lack on exploration ability of new domains.ML conversely provides little explicit physical insight whilst yielding a stronger pre-diction accuracy for large-scale data.Hybrid PM/ML approaches provide solutions maximising accuracy,while leading to a clearer physical picture and the desired properties.     

Accelerating materials discovery using machine learning

The discovery of new materials is one of the driving forces to promote the development of modern society and technology innovation,the traditional materials research mainly depended on the trial-and-error method,which is time-consuming and laborious.Recently,machine learning (ML) methods have made great progress in the researches of materials science with the arrival of the big-data era,which gives a deep revolution in human society and advance science greatly.However,there exist few systematic generalization and summaries about the applications of ML methods in materials science.In this review,we first provide a brief account of the progress of researches on materials science with ML employed,the main ideas and basic procedures of this method are emphatically introduced.Then the algorithms of ML which were frequently used in the researches of materials science are classified and compared.Finally,the recent meaningful applications of ML in metal materials,battery materials,photovoltaic materials and metallic glass are reviewed.     

Effect of stress profile on microstructure evolution of cold-drawn commercially pure aluminum wire analyzed by finite element simulation

The evolution of microstructure in the drawing process of commercially pure aluminum wire (CPAW) does not only depend on the nature of materials, but also on the stress profile. In this study, the effect of stress profile on the texture evolution of the CPAW was systematically investigated by combining the numerical simulation and the microstructure observation. The results show that the tensile stress at the wire center promotes the formation of <111> texture, whereas the shear stress nearby the rim makes little contribution to the texture formation. Therefore, the <111> texture at the wire center is stronger than that in the surface layer, which also results in a higher microhardness at the center of the CPAW under axial loading.     

A Change-Point Approach for Phase-I Analysis in Multivariate Profile Monitoring and Diagnosis

Process monitoring and fault diagnosis using profile data remains an important and challenging problem in statistical process control (SPC). Although the analysis of profile data has been extensively studied in the SPC literature, the challenges associated with monitoring and diagnosis of multichannel (multiple) nonlinear profiles are yet to be addressed. Motivated by an application in multioperation forging processes, we propose a new modeling, monitoring, and diagnosis framework for phase-I analysis of multichannel profiles. The proposed framework is developed under the assumption that different profile channels have similar structure so that we can gain strength by borrowing information from all channels. The multidimensional functional principal component analysis is incorporated into change-point models to construct monitoring statistics. Simulation results show that the proposed approach has good performance in identifying change-points in various situations compared with some existing methods. The codes for implementing the proposed procedure are available in the supplementary material.     

基于遗传算法和自适应的平面线轮廓度误差评定方法

 为了解决在测量平面线轮廓度中由于存在被测轮廓与其测量基准间存在位置误差而影响评定精度的问题，提出了一种基于遗传算法和自适应的计算平面线轮廓度误差的新方法。该方法满足最小条件原理，它利用样条插值函数拟合理论轮廓，并在评定过程中能自动地实现被测轮廓与理论轮廓之间的适应性调整，从而能够分离并消除被测轮廓与其测量基准之间的位置误差对轮廓误差评定结果的影响，在遗传优化中获得全局最优解。实例计算验证了这一结果。这种算法简单明确，具有精度高、收敛速度快、易于计算机程序实现、易于推广应用等特点。     

A ZnS/CaZnOS Heterojunction for Efficient Mechanical-to-Optical Energy Conversion by Conduction Band Offset

Actively collecting the mechanical energy by efficient conversion to other forms of energy such as light opens a new possibility of energy-saving, which is of pivotal significance for supplying potential solutions for the present energy crisis. Such energy conversion has shown promising applications in modern sensors, actuators, and energy harvesting. However, the implementation of such technologies is being hindered because most luminescent materials show weak and non-recoverable emissions under mechanical excitation. Herein, a new class of heterojunctioned ZnS/CaZnOS piezophotonic systems is presented, which displays highly reproducible mechanoluminescence (ML) with an unprecedented intensity of over two times higher than that of the widely used commercial ZnS (the state-of-the-art ML material). Density functional theory calculations reveal that the high-performance ML originates from efficient charge transfer and recombination through offset of the valence and conduction bands in the heterojunction interface region. By controlling the ZnS-to-CaZnOS ratio in conjunction with manganese (Mn²⁺) and lanthanide (Ln³⁺) doping, tunable ML across the full spectrum is activated by a small mechanical stimulus of 1 N (10 kPa). The findings demonstrate a novel strategy for constructing efficient ML materials by leveraging interface effects and ultimately promoting practical applications for ML.     

Effective behavior of porous elastomers containing aligned spheroidal voids

The theoretical need to recognize the link between the basic microstructure of nonlinear porous materials and their macroscopic mechanical behavior is continuously rising owing to the existing engineering applications. In this regard, a semi-analytical homogenization model is proposed to establish an overall, continuum-level constitutive law for nonlinear elastic materials containing prolate/oblate spheroidal voids undergoing finite axisymmetric deformations. The microgeometry of the porous materials is taken to be voided spheroid assemblage consisting of confocally voided spheroids of all sizes having the same orientation. Following a kinematically admissible deformation field for a confocally voided spheroid, which is the basic constituent of the microstructure, we make use of an energy-averaging procedure to obtain a constitutive relation between the macroscopic nominal stress and deformation gradient. In this work, both prolate and oblate voids are considered. As a numerical example, we study macroscopic nominal stress components for a hyperelastic porous material consisting of a neo-Hookean matrix and prolate/oblate voids subjected to 3-D and plane strain dilatational loadings. In this numerical study, the relation between the relevant microstructural variables (i.e., initial porosity and void aspect ratio) for a rather large range of applied stretch is put into evidence for two types of loading. Finally, a finite element (FE) simulation is presented, and the homogenization model is assessed through comparison of its predictions with the corresponding FE results. The illustrated agreement between the results demonstrates a good accuracy of the model up to rather large deformations.     

基于线结构光的叶片型面特征检测方法研究

叶片是燃气轮机和航空发动机的核心部件,随着我国两机行业的高速发展,对于叶片在研发、生产和维修等方面全生命周期检测的要求不断提高。光学测量是目前叶片三维形貌高效检测的新手段,但是在测量精度等方面相比传统三坐标测量机仍存在一定局限。该文提出一种基于线结构光的叶片型面特征检测方法,设计开发一套四自由度检测平台,针对基于标定物的平台位姿校准和数据采集方法开展研究,并以某气轮机导向叶片为检测对象进行实验测试。测试数据与精密三坐标测量机实测数据的对比结果表明:型面型线轮廓度偏差在±0.02 mm以内,截面主要特征参数偏差均在±0.018 mm以内,该文面向叶片型面实际检测需求的方法可行。     

Machine Learning Interatomic Potentials as Emerging Tools for Materials Science

Atomic‐scale modeling and understanding of materials have made remarkable progress, but they are still fundamentally limited by the large computational cost of explicit electronic‐structure methods such as density‐functional theory. This Progress Report shows how machine learning (ML) is currently enabling a new degree of realism in materials modeling: by “learning” electronic‐structure data, ML‐based interatomic potentials give access to atomistic simulations that reach similar accuracy levels but are orders of magnitude faster. A brief introduction to the new tools is given, and then, applications to some select problems in materials science are highlighted: phase‐change materials for memory devices; nanoparticle catalysts; and carbon‐based electrodes for chemical sensing, supercapacitors, and batteries. It is hoped that the present work will inspire the development and wider use of ML‐based interatomic potentials in diverse areas of materials research.     

纳米固体材料的界面及其对性能的影响

纳米固体具有一系列特殊的性能。一般认为这些特性在很大程度上由纳米固体的特殊徽结构一很高浓度的缺陷（界面组元、空隙、位错等）、特殊的界面结构及缺陷组成所决定的。讨论了纳米固体的界面组成、结构及其影响界面结构的主要因素，评述了纳米固体材料的界面对其机械性能的影响，简要讨论了界面变化引起这些性能变化的机理。     

Carbon seal materials with superior mechanical properties and fine-grained structure fabricated by a new process

Carbon seal material was prepared using a new liquid mixing process of the raw materials, which had not only lowered open porosity but also provided excellent mechanical properties, especially higher compressive strength (210 MPa) compared with the material prepared by the conventional method. Scanning electron microscopy (SEM) results show that such material has a fine-grained structure as well as little pore diameter. The influence of the manufacturing procedure of the materials on the performance and microstructure is investigated. In addition, correlations between properties and microstructure are also discussed.     

A three-step renumbering procedure for high-order finite element analysis

An efficient renumbering method for high-order finite element models is presented. The method can be used to reduce the profile and wavefront of a coefficient matrix arising in high-order finite element computation. The method indirectly performs node renumbering and involves three main steps. In the first step, nodes at corners of the elements are numbered using an existing renumbering algorithm. In the second step, elements are numbered in an ascending order of their least new corner node numbers. Finally, based on the new element numbers, both corner and non-corner nodes are renumbered using an algorithm that simulates the node elimination procedure in a frontal solution method. The method is compared to the algorithms that directly perform node renumbering. The numerical results indicate that the three-step algorithm presented here is an order of magnitude faster and the resulting renumbering produces excellent profile and wavefront characteristics of the coefficient matrix. © 1998 John Wiley & Sons, Ltd.