首页期刊导航|Journal of Materials Processing Technology
期刊信息/Journal information
Journal of Materials Processing Technology
Elsevie
Journal of Materials Processing Technology

Elsevie

0924-0136

Journal of Materials Processing Technology/Journal Journal of Materials Processing TechnologyISTPSCIEI
正式出版
收录年代

    Towards high-performance deep learning models in tool wear classification with generative adversarial networks

    Molitor, Dirk AlexanderKubik, ChristianBecker, MarcoHetfleisch, Ruben Helmut...
    14页
    查看更多>>摘要：Image-based deep learning (DL) applications have been increasingly applied in manufacturing processes in recent years and enable real-time decisions to be derived on the basis of raw image data. However, there are major obstacles that prevent a widespread application. The generation of extensively labeled datasets is associated with costly experiments, long downtimes and demanding measurement setups. Thus, users are usually faced with a trade-off problem, where they have to weigh between the profitability and the performance of their DL algorithms. For this reason, this paper demonstrates how this obstacle can be tackled through the application of data augmentation (DA) techniques. The investigations are based on an image dataset from a tool wear classification task in a blanking process, in which a pre-trained convolutional neural network (CNN) identifies correlations between cutting punch radii and workpiece images. It is shown how synthesized images in combination with transfer learning (TL) models affect classification accuracies using basic image manipulation, different types of generative adversarial networks (GAN) and their hybrid application, varying the number of available images in the training dataset between 160 and 4800. These augmentation techniques lead to accuracy improvements of up to 18%, although their effectiveness depends heavily on the amount of image data available.
      原文链接:
    • NSTL
    • Elsevier

    Fatigue behavior and modeling of additively manufactured IN718: The effect of surface treatments and surface measurement techniques

    Lee, SeungjongShao, ShuaiWells, Douglas N.Zetek, Miroslav...
    15页
    查看更多>>摘要：In this study, the effect of various surface treatments, including sand-blasting, drag-finishing, turning, grinding, and grinding + drag-finishing, on surface roughness and fatigue properties of laser beam powder bed fused Inconel 718 was examined. The surface roughness values obtained from two surface measurement techniques, i.e., optical microscopy and X-ray computed tomography, were compared. Both surface measurement techniques consistently indicated that all surface treatments led to improvements in surface roughness, although optical microscopy was influenced by surface glares and overestimated the surface roughness values of drag-finished specimens. Accordingly, all surface treatments also led to improvement in fatigue resistance with sandblasting and drag-finishing yielding the least while grinding + drag-finishing causing the most. Notably, only the cracks of grinding + drag-finished specimens initiated from crystallographic facets while those in other conditions were surface initiated. Furthermore, by treating the surface valleys as micro notches, an effective fatigue notch factor model using a hybrid surface roughness metric that incorporates several standard surface roughness parameters was shown to correlate the fatigue lives of 94 % of specimens with various surface conditions within a scatter band of three.
      原文链接:
    • NSTL
    • Elsevier

    Effects of rescanning parameters on densification and microstructural refinement of 316L stainless steel fabricated by laser powder bed fusion

    Liang, AnqiPey, Khee SiangPolcar, TomasHamilton, Andrew R....
    13页
    查看更多>>摘要：A challenge with microstructural control and refinement in laser powder bed fusion (LPBF) is maintaining high density when choosing parameters for desired microstructures. Rescanning during LPBF has been reported to improve densification and decrease surface roughness for many different alloys. However, little has been reported regarding the effects of locally rescanning with varying processing parameters on sub-grain cell size refinement for 316L stainless steel (SS). This study presents a novel solution to enable high densification with microstructural control in 316L SS by using a set of initial scanning parameters to achieve densification and a different set of rescanning parameters to refine the microstructure. Results showed that rescanning resulted in heterogeneous microstructure with coarse cell size of 0.84 mu m and locally refined cell size of 0.35 mu m, while maintaining a high level of densification (99.96 %), therefore enabling potential variations in component strength and hardness. The spatial distribution of local microstructure refinement was dictated by the melt pool dimensions of initial scanning and rescanning relative to the powder layer thickness. To better understand the link between LPBF process parameters and microstructure, the Wilson-Rosenthal equation was used to predict cooling rate (G x R) and correlate with sub-grain cell size. Such variation in properties may be useful for applications requiring parts with hardened surfaces, or localized strengthening at stress concentrations and sites of expected failure.
      原文链接:
    • NSTL
    • Elsevier

    Direct-Ink-writing of liquid metal-graphene-based polymer composites: Composition-processing-property relationships

    Tandel, RuchiraGozen, B. Arda
    10页
    查看更多>>摘要：This paper presents a study of the solvent-based precursors (inks) derived from composites of polyethylene oxide, graphene flakes and micro-scale spherical Eutectic Gallium Indium (EGaIn) fillers and their processing through direct-ink-writing (DIW). The presented studies focus on the influence of EGaIn fillers on ink rheology, DIW process mechanisms as well as electrical conductivity of the printed structures. The results show that EGaIn fillers vary the ink rheology towards a more elastic behavior with lower extensional viscosity. This leads to ink filaments capable of withstanding large extensional strains during DIW, forming continuous prints even when printing speed is higher than ink flow speed, and producing features with line width smaller than the nozzle diameter. Electrical conductivity of the prints reduces with increasing strain due to the deformation of the liquid EGaIn fillers along the printing direction. These findings can be utilized to control the DIW process and the properties of the conductive polymer composites.
      原文链接:
    • NSTL
    • Elsevier

    Deep DIC: Deep learning-based digital image correlation for end-to-end displacement and strain measurement

    Yang, RuLi, YangZeng, DanielleGuo, Ping...
    15页
    查看更多>>摘要：Digital image correlation (DIC) has become an industry standard to retrieve accurate displacement and strain measurement in tensile testing and other material characterization. Though traditional DIC offers a high precision estimation of deformation for general tensile testing cases, the prediction becomes unstable at large deformation or when the speckle patterns start to tear. In addition, traditional DIC requires a long computation time and often produces a low spatial resolution output affected by filtering and speckle pattern quality. To address these challenges, we propose a new deep learning-based DIC approach - Deep DIC, in which two convolutional neural networks, DisplacementNet and StrainNet, are designed to work together for end-to-end prediction of displacements and strains. DisplacementNet predicts the displacement field and adaptively tracks a region of interest. StrainNet predicts the strain field directly from the image input without relying on the displacement prediction, which significantly improves the strain prediction accuracy. A new dataset generation method is developed to synthesize a realistic and comprehensive dataset, including the generation of speckle patterns and the deformation of the speckle image with synthetic displacement fields. Though trained on synthetic datasets only, Deep DIC gives highly consistent and comparable predictions of displacement and strain with those obtained from commercial DIC software for real experiments, while it outperforms commercial software with very robust strain prediction even at large and localized deformation and varied pattern qualities. In addition, Deep DIC is capable of real-time prediction of deformation with a calculation time down to milliseconds.
      原文链接:
    • NSTL
    • Elsevier

    In-situ enhanced laser absorption in aqueous transition metal salt solution enables high-quality backside wet etching of optical glass by near-infrared lasers

    Long, JiangyouLi, YuxinEliceiri, Matthew H.Lai, Qing...
    10页
    查看更多>>摘要：A laser-induced backside wet etching technique using near-infrared laser sources (N-LIBWE) has been proposed for high-quality micromachining of optical glass. Compared with conventional LIBWE using organic dye solutions with high absorption coefficients (similar to 10(4)-10(5) cm(-1)) as the liquid absorbent, the transition metal salt solution used in N-LIBWE shows the absorption coefficient of only 10(-1)-10(1) cm(-1). Moreover, as we proved in this study, the etching rate of N-LIBWE is independent of the absorptivity of the liquid absorbent. How the etching occurs during N-LIBWE is still controversial. Herein, interface images with a temporal resolution of 4 ns after the start of laser irradiation are captured by a stroboscopic shadowgraphy system. Afterward, the density, pressure, and temperature distribution in the solution are calculated by a physics-based two-dimensional hydrodynamic model to explain the observed interface images. We prove that the localized laser absorption is enhanced with laser irradiation because of the in-situ formation of solid substances. It is the properties of the formed solid products instead of the normal absorption coefficient of the solution that determines the final interface status. We further prove that the formed solid substances in aqueous CuSO4 solutions are primarily amorphous nano clusters, which benefit the enhancement of laser absorption and tend to attach to the transparent substrates, resulting in the incubation effect in N-LIBWE. Our results provide systematic mechanisms explaining the NLIBWE process, which also benefit the deep understanding of other liquid-assisted laser materials processing technologies.
      原文链接:
    • NSTL
    • Elsevier

    Investigation on joining high borated stainless steels through electron beam welding technology

    Wang, Zhao-JieYin, Fan-jingLi, Yong-WangXie, Guang-Ming...
    13页
    查看更多>>摘要：High borated stainless steels (HBSSs) are widely used in nuclear field because of excellent neutron-shielding ability. However, coarse network-like borides are prone to be formed in weld joints during welding process, deteriorating the ductility seriously. In this work, electron beam (EB) welding was adopted to join-3.5 mm thick HBSSs containing 0.30-2.12 wt. % boron. Benefiting from high energy density, fine gamma-Fe matrix and network-like borides were obtained in fusion zone (FZ) and partially molten zone (PMZ) of weld joints. The effects of boron content, heat input and post weld heat treatment (PWHT) on the microstructure, mechanical properties and fracture behaviour of weld joints were investigated. As boron content increased, narrower PMZ was observed in weld joints owing to smaller solidification temperature range. The volume fraction and size of borides in PMZ/FZ raised, enhancing the strength and decreasing the ductility. Meanwhile, the fracture location of weld joints was closer to the interface of PMZ and base metal (BM) during deformation. Higher heat input brought a wider PMZ and larger penetration depth, but hardly changed the size and distribution of borides. Interestingly, it caused a steep slope of interfacial line between PMZ and BM, leading to the fracture along the interface with low ductility. After PWHT, the borides in PMZ/FZ were spheroidized, and the stress localization during deformation was weakened. Consequently, excellent weld joints exhibiting similar mechanical properties with base steels were produced. This systematic work provided a reference for welding other materials containing coarse and brittle secondary phases.
      原文链接:
    • NSTL
    • Elsevier

    Machine learning-based modeling of the coupling effect of strain rate and temperature on strain hardening for 5182-O aluminum alloy

    Shang, HongchunWu, PengfeiLou, YanshanWang, Jizhen...
    22页
    查看更多>>摘要：This research characterizes the dynamic hardening behavior of an aluminum alloy sheet of 5182-O for the coupling effect of strain rate and temperature. Tests are carried out for dogbone specimens at different loading conditions to experimentally characterize the strain rate hardening and thermal softening effect for the alloy. The behaviours are then modeled by the Johnson-Cook, Zerilli-Armstrong and Lim-Huh models. In addition, the FEAfriendly polynomial model and artificial neural network (ANN) model are used to describe the highly non linearity and coupling of strain hardening. Factors affecting ANN predicting accuracy and numerical computing efficiency are comprehensively studied including network structure, parameter settings and optimization algorithms. All the analytical and ANN models are also implemented into ABAQUS/Explicit to numerically compute the reaction force of tensile tests of dogbone specimens. The strain hardening curves are predicted by the analytical and ANN models for the comparison with experimental measurements to evaluate their performance. The experimental results show that the strain rate is slightly negative at room temperature, while the strain rate effect turns to be positive as temperature rises. The comparison of the flow curves between prediction and experiments reveals that the coupling effect is reasonably illustrated by the proposed polynomial model and the ANN model illustrates the flow curves with the dramatically much better accuracy than all the other models. The numerically predicted reaction forces prove that the ANN model accurately illustrates the load capability with the best agreement among the models studied in this research. The numerical computation also shows that the numerical computation efficiency of the ANN model is slightly reduced compared with analytical models, but the reduction is not much and worthwhile compared with the high accuracy of ANN.
      原文链接:
    • NSTL
    • Elsevier

    A theoretical model to predict the anisotropic characteristics in grinding of diamond conical indenter

    Wu, LiqiangZhang, HaijunZong, WenjunDu, Kai...
    15页
    查看更多>>摘要：Mechanical grinding of a high-precision diamond conical indenter is difficult due to the strong grinding removal rate anisotropy. In the existing literature, theoretical models have been established to predict the strength anisotropy of diamond on the (100), (110) and (111) planes, which provide a theoretical basis for the fabrication of pyramid diamond indenters, such as Berkovich, Vickers or Knoop indenters. However, investigations on the anisotropic characteristics in processing of diamond curved surface are rarely carried out. To solve the anisotropy problem in grinding of diamond conical indenter, this work contributes a theoretical model to predict the anisotropic characteristics in grinding of diamond cone face, and provides an effective method to optimize the grinding direction for enhancing the fabrication precision. Based on a proposed grinding-ability factor model, a general method is developed to analyze the grinding ability along different directions according to the formation process of the cone face. The results suggest that the optimized mechanical grinding direction can be determined by minimizing the standard deviation of the grinding-ability factors of the crystal planes. The proposed method is validated by grinding experiments on indenters, with the <100> or <111> orientation along the indenter axis. The experimental observations confirm that the optimized grinding direction can weaken the pyramid phenomenon, yielding a roundness error of 0.12 mu m. In addition, the grinding removal rate-dependent roughness petal phenomenon is suppressed, which considerably enhances the conical surface quality.
      原文链接:
    • NSTL
    • Elsevier

    A novel low-damage and low-abrasive wear processing method of C-f/SiC ceramic matrix composites: Laser-induced ablation-assisted grinding

    Zhou, KunXu, JiayuXiao, GuijianHuang, Yun...
    15页
    查看更多>>摘要：Continuous fiber-reinforced ceramic matrix composites are promising materials for advanced hot end components and safety-critical components of aeronautics and astronautics; however, the difficult-to-process problem of these materials has been one of the technical bottlenecks restricting their high-performance service. The present work proposed a novel high-efficiency, low-damage, and low-abrasive wear processing method based on the chemical properties of the materials, which used lasers to ablate workpieces before grinding-laser-induced ablation-assisted grinding (LIAAG). The material removal mechanism of Cf/SiC composites was investigated by single-grain scratching tests; the grinding performances of LIAAG in terms of grinding force, grinding temperature, surface integrity, abrasive wear, grinding chips were evaluated by a contrast abrasive belt grinding test. It was found that Cf/SiC composites were chemically transformed into relatively loose and homogeneous ablation products (SiO2 and recrystallized SiC) at high laser ablation temperatures, which were easier to remove during grinding. In this case, surface morphologies displayed the microfracture and crushing of carbon fibers and SiC matrixes, and the grinding-induced damages, such as macro fracture, fiber pull-out, interface debonding, were significantly contained. Abrasive belt was primarily worn in micro-adhesion and micro-abrasion, rather than cleavage fracture and the fall-off in traditional grinding. Under optimum parameters, the grinding force, grinding temperature, and average surface roughness of belt grinding were reduced by 47 %, 40 %, and 26 %, respectively. Consequently, the surface integrity was improved greatly, and the abrasive wear was reduced significantly. The work provides a vital high-performance processing method for ceramic matrix composites components.
      原文链接:
    • NSTL
    • Elsevier