Search

To establish the finite element model of T12 and L2 (T12-L2) pedicle screw fixation for severe L1 burst fracture, and quantitatively simulate and analyze the screw stress and vertebral displacement in different degrees of L1 posterior upper wall fracture (PUWF), and evaluate whether PUWF degree is a risk factor for fixation failure.The data of 6 healthy volunteers were used to establish a finite element model of T12-L2 pedicle screw fixation for type A severe burst fractures. The stress and displacement of the conventional and Schanz pedicle screws for the different degrees of PUWF (including the anterior upper wall of the vertebral canal and the bipedicle) were evaluated.The maximum stress and L1 displacement of conventional and Schanz pedicle screws were positively correlated with the severity of the PUWF (P0.05). During anterior flexion, the maximum stress of conventional pedicle screws for 70% type I were 538.3±59.75MPa and the maximum stress of Schanz pedicle screws for 90% type Ⅱ, 90% type Ⅲ and 70% type IV fractures were close to the fatigue threshold. The maximum stress during anterior flexion were significantly higher than those during posterior extension, bending and rotation (P0.05).The posterior upper wall fracture of vertebral body (VB) of type A burst fracture is not an independent risk factor for the failure of short-segment pedicle screw fixation (SSPSF). Anterior flexion of type A fractures combined with severe PUWF of VB was a risk factor for the failure of SSPSF.

WAAM has been proven a promising alternative to fabricate medium and large scale metal parts with a high depositing rate and automation level. However, the production quality may deteriorate due to the poor deposited layer surface quality. In this paper, a laser sensor based surface roughness measuring method was developed for WAAM. To improve the surface integrity of deposited layers by WAAM, different machine learning models, including ANFIS, ELM and SVR, were developed to predict the surface roughness. Furthermore, the ANFIS model was optimized by GA and PSO algorithms. Full factorial experiments were conducted to obtain the training data, and the K-fold Cross-validation strategy was applied to train and validate machine learning models. The comparison results indicate that GA–ANFIS has superiority in predicting surface roughness. The RMSE, $$ R^{2} $$ , MAE and MAPE for GA–ANFIS were 0.0694, 0.93516, 0.0574, 14.15% respectively. This study could also provide inspiration and guidance for surface roughness modelling in multipass arc welding and cladding.

Background To evaluate the biomechanical properties of short-segment Schanz screw implanted in an oblique downward direction for the treatment of lumbar 1 burst fracture using a finite element analysis. Methods The Universal Spine System (USS) fixation model for adjacent upper and lower vertebrae (T12 and L2) of lumbar 1 vertebra burst fracture was established. During flexion/extension, lateral bending, and rotation, the screw stress and the displacement of bone defect area of the injured vertebrae were evaluated when the downward inserted angle between the long axis of the screws and superior endplate of the adjacent vertebrae was set to 0° (group A), 5° (group B), 10° (group C), and 15°(group D). There were 6 models in each group. Results There were no significant differences in the maximum screw stress among all the groups during flexion/extension, lateral bending, and rotation (P > 0.05). There were no significant differences in the maximum displacement of the bone defect area of the injured vertebrae among all the groups during flexion/extension, lateral bending, and rotation (P > 0.05). Conclusion Short-segment Schanz screw implanted in an oblique downward direction with different angles (0°/parellel, 5°, 10°, and 15°) did not change the maximum stress of the screws, and there was a lower risk of screw breakage in all groups during flexion/extension, lateral bending, and rotation. In addition, the displacement of the injured vertebra defect area had no significant changes with the change of angles.

Wire arc additive manufacturing (WAAM) is an emerging technology in the manufacturing industry, which uses a welding arc as an energy source to fuse metal wire and deposit layer by layer. In order to promote its manufacture precision, stability, and repeatability, it’s crucial to develop a feedback control strategy for WAAM. This research implements vision-based feedback control for the layer width during the WAAM process. A WAAM system is developed using a robot and CMT welder with a visual sensing system. The dynamics of the layer width in WAAM process is modeled experimentally. An ARX dynamic model is built. Based on this model, a model predictive control (MPC) strategy is derived to regulate the WAAM process. Feedback control experiments were conducted to verify the tracking and robustness performance of the proposed MPC algorithm.

Wire arc additive manufacturing technology (WAAM) has become a very promising alternative to high-value large metal components in many manufacturing industries. Due to its long process cycle time and arc-based deposition, defect monitoring, process stability and control are critical for the WAAM system to be used in the industry. Although major progress has been made in process development, path slicing and programming, and material analysis, a comprehensive process monitoring, and control system are yet to be developed. This paper aims to provide an in-depth review of sensing and control design suitable for a WAAM system, including technologies developed for the generic Arc Welding process, the Wire Arc Additive Manufacturing process and laser Additive Manufacturing. Particular focus is given to the integration of sensor-based feedback control, and how they could be implemented into the WAAM process to improve its accuracy, reliability, and efficiency. The paper concludes by proposing a framework for sensor-based monitoring and control system for the GMAW based WAAM process. This framework provides a blueprint for the monitoring and control strategies during the WAAM process and aims to identify and reduce defects using information fusion techniques.

Wire arc additive manufacturing (WAAM) is a Direct Energy Deposition (DED) technology, which utilize electrical arc as heat source to deposit metal material bead by bead to make up the final component. However, issues like the lack of assurance in accuracy, repeatability and stability hinder the further application in industry. Therefore, a Model Free Adaptive Iterative Learning Control (MFAILC) algorithm was developed to be applied in WAAM process in this study. The dynamic process of WAAM is modelled by adaptive neuro fuzzy inference system (ANFIS). Based on this ANFIS model, simulations are performed to demonstrate the effectiveness of MFAILC algorithm. Furthermore, experiments are conducted to investigate the tracking performance and robustness of the MFAILC controller. This work will help to improve the forming accuracy and automatic level of WAAM.

As an advanced and highly efficient welding method, Keyhole Tungsten Inert Gas (keyhole TIG) welding has drawn wide interests from the manufacturing industry. In order to improve its manufacturing quality and automation level, it’s necessary to develop an online monitoring system for the keyhole TIG welding process. This study developed a visual monitoring system, which utilized an HDR welding camera to monitor the welding pool and keyhole during keyhole TIG welding process. A state of the art Convolutional neural network (Resnet) was developed to recognize different welding states, including good weld, incomplete penetration, burn through, misalignment and undercut. In order to improve the diversity of training dataset, image augmentation was performed. To optimize the training process, a metric learning strategy of center loss was introduced. Furthermore, visualization methods, including guided Grad-CAM, feature map and t-SNE were applied to understand and explain the effectiveness of deep learning process. This study will lay a solid foundation for the development of on-line monitoring system of keyhole TIG.

In this study, rice protein was used as raw material to explore the effects of γ-irradiation treatment doses (0, 0.5, 1, 2, 3, 5 kGy) on the physicochemical properties of rice protein (particle size, zeta potential, secondary structure, scanning electron microscope microstructure), surface hydrophobicity (H0), thermal stability), functional properties (solubility, water and oil retention, emulsification) and sensory quality. The results show that when the γ-irradiation dose is 2 kGy, the average particle size of rice protein is the smallest, the absolute value of the potential is the highest 33.58 mV, the content of β-sheets in the secondary structure is at least 31.16 ± 0.16, and the content of random curl is at most 14.56 ± 0.06, the surface of the microstructure is rough and the degree of pore depression is the deepest, the highest H0 is 160.45 ± 2.98, the minimum denaturation temperature (Td) and enthalpy (△H) are 70.49 ± 0.05 °C and 1.30 ± 0.01 J/g, which shows that γ-irradiation treatment can be significant affect the physicochemical properties of rice protein. When the irradiation dose is 2 kGy, the highest solubility of rice protein is 69.18 ± 1.07%, and the highest water and oil holding capacity are 5.89 ± 0.08 g/g and 3.45 ± 0.04 g/g, respectively. The highest emulsification activity and emulsification stability are 45.65 ± 1.26 m2/g and 208.33 ± 4.79 min, which shows that γ-irradiation treatment can improve the functional properties of protein. When the irradiation dose was less than 5 kGy, the sensory quality of rice protein was not significantly affected. The research results provide a theoretical basis for the deep processing and value-added utilization of rice protein by γ-irradiation technology.

Wire-arc additive manufacturing (WAAM) technology has been widely recognized as a promising alternative for fabricating large-scale components, due to its advantages of high deposition rate and high material utilization rate. However, some anomalies may occur during the deposition process, such as humping, spattering, and robot suspend. this study proposed to apply Deep Learning in the visual monitoring to diagnose different anomalies during WAAM process. The melt pool images of different anomalies were collected for training and validation by a visual monitoring system. The classification performance of several representative CNN architectures, including ResNet, EfficientNet, VGG-16 and GoogLeNet, were investigated and compared. The classification accuracy of 97.62%, 97.45%, 97.15% and 97.25% was achieved by each model. The results proved that the CNN models are effective in classifying different types of melt pool images of WAAM. Our study is applicable beyond WAAM and should benefit other additive manufacturing or arc welding techniques.

Geometric properties of material deposited by the wire arc additive manufacturing (WAAM) process often deviates from desired setpoints. To improve the accuracy and repeatability of the WAAM process, an effective control strategy to maintain desired deposition geometry that operates robustly under various welding conditions is required. In this work, a control strategy utilizing multi-input multi-output (MIMO) model-predictive control (MPC) is presented. This approach, based on linear autoregressive (ARX) modelling, aims to improve the accuracy and flexibility of deposited bead geometry in the WAAM process. The MPC controller updates welding parameters between successive layers by minimizing a cost function based on sequences of input variables. Measurements of deposited bead geometry are made by laser scanner and input to the linear ARX model, which then makes future bead geometry predictions. Weighting coefficients of the ARX model are trained iteratively throughout the manufacturing process. Experimental results show that the derived control strategy can reduce fluctuations in a part's height by 400% and maintain the fluctuation within an acceptable range. In addition, the fluctuations in bead width along a single weld seam was also improved by more than 50%.

In more recent times, research on various aspects of the Wire Arc Additive Manufacturing (WAAM) process has been conducted, and efforts into monitoring the WAAM process for defect identification have increased. Rapid and reliable monitoring of the WAAM process is a key development for the technology as a whole, as it will enable components produced by the process to be qualified to relevant standards and hence be deemed fit for use in applications such as those found in the aerospace or naval sectors. Intelligent algorithms provide inbuilt advantages in processing and analysing data, especially for the large data sets generated during the long manufacturing cycles. Interdisciplinary engineering (IDE) furnishes a concept integrating computer science and industrial system manufacturing engineering together to treat large amounts of process monitoring data. In this work, a WAAM process monitoring and defect detection system integrating intelligent algorithms is presented. The system monitors welding arc current and voltage signals produced by the WAAM process and makes use of a support vector machine (SVM) learning method to identify disturbances to the welding signal which indicate the presence of potential defects. The incremental machine learning models developed in this work are trained via statistical feature analysis of the welding signals and a novel quality metric that improves detection rates is also presented. The incremental learning approach provides an efficient means of detecting welding-based defects, as it does not require large quantities of data to be trained to an operational level (addressing a major drawback of other machine learning methods). A case study is presented to validate the developed system, results show that it was able to detect a set of defects with a success rate greater than 90% F1-score.

Cu2ZnSnS4 is a promising and noticeable material for the absorber layer of thin film solar cells. Suitable off-stoichiometic composition of pure kesterite phase is profitable for the photoelectron properties, but such compositions tend to cause the appearance of secondary phases. The stable composition range of pure kesterite phase is difficult to confirm by experiment. In this paper, calculation models of single kesterite phase and other possible phases in Cu-Zn-Sn-S alloy system were constructed and the formation energies E were calculated based on first-principle method. The stability of single kesterite phase with off-stoichiometric compositions was deduced by comparing its E with that of different phase combinations in Cu-Zn-Sn-S alloy system. Following conclusions were drawn. When Cu/(Zn+Sn) ≤ 0.778, single kesterite phase is not as stable as any of phase combinations because of its smallest E, so secondary phases definitely appear. When Cu/(Zn+Sn) = 0.778–0.882, the secondary phases should also appear on equilibrium condition, while they may not appear on non-equilibrium condition because the E difference is relative small. When Cu/(Zn+Sn) = 1.000–0.882, single kesterite phase should be the stablest phase and the secondary phases should not appear on equilibrium condition. All experimental results we could find are in accordance with the calculated kesterite phase stable range. Such a range can provide a profitable guidance for the composition design of Cu-Zn-Sn-S alloys for thin film solar cells. The calculation model and method can be further extended for the prediction of the phase stability in other alloy systems.

An approach based on millisecond pulsed laser is proposed for thermal fatigue property evaluation in this paper. Cyclic thermal stresses and strains within millisecond interval are induced by complex and transient temperature gradients with pulsed laser heating. The influence of laser parameters on surface temperature is studied. The combination of low pulse repetition rate and high pulse energy produces small temperature oscillation, while high pulse repetition rate and low pulse energy introduces large temperature shock. The possibility of application is confirmed by two thermal fatigue tests of compacted graphite iron with different laser controlled modes. The developed approach is able to fulfill the preset temperature cycles and simulate thermal fatigue failure of engine components.

For hot metal forming at high temperatures, lubricant is required to reduce friction and wear, as well as improve the product surface quality. Inorganic compound has been regarded as a candidate to replace the traditional oil/water lubricant for the forming process. Friction and wear behaviour of three inorganic compounds, including sodium polyphosphate, sodium carbonate, sodium borate were evaluated under ball-on-disc tests at high temperature for high speed steel/silicon steel counterparts to simulate the hot rolling process. The results show that all the three compounds can reduce the friction and wear. Sodium polyphosphate is more corrosive with the oxide scale than others, which is believed to contribute to the descaling of red scale of silicon (Si) steel. Mechanisms by which the inorganic compounds improve the tribological behaviours were discussed here. Simulation results indicate that the molecules of sodium polyphosphate and sodium borate show a much higher binding energy on the iron oxide surface than that of sodium carbonate. The binding energy is mainly contributed from the ionic and the covalent interactions. The ionic part comes from the interaction between the Na+ cations and the oxide surface, while the covalent part comes from the chemical bonds between Fe and oxygen of the lubricants. Both of them contribute to the adhesion of the lubricants on the steel surface and are vital to the friction and wear performance of the lubricants.

As a new emerging technology, wire arc additive manufacturing (WAAM) has attracted extensive interests from both academia and industry during recent years. WAAM uses welding arc as an energy source to fuse metal wire and deposit layer by layer, which provides the advantages of freeform deposition. In order to improve its manufacture precision, stability and repeatability, it is necessary to develop sensing and control strategy for WAAM process. This research develops a passive visual sensing system for a robotic WAAM system. A new deep learning technique (Mask R-CNN) is proposed to detect and segment the melt pool area, and the width of melt pool can be measured based on the coordinate of the bounding rectangle. The pseudo-random ternary (PRT) signals were used to stimulate the WAAM process, and the corresponding width can be measured by the Mask R-CNN. Based on the width data and corresponding PRT input, a dynamic model of adaptive neuro-fuzzy inference system was built for the WAAM process.

Adequate lubrication for the sliding contacts is responsible for the energy saving and long service life. In this work, load-dependence of tribological behaviours of sodium carbonate coating on the stainless steel was evaluated by ball-on-disc friction test at high temperatures. The results suggested that this sodium carbonate coating showed promising lubrication even under high applied load of 1800 N (Hertz pressure is 6.14 GPa). Apart from the improvement of lubricity, it can also inhibit adhesive wear that transfer oxide scale from the disc to the ball under dry sliding condition. The desirable lubrication and load-carrying capability were attributed to the frictional heat induced in-situ formed sodium carbonate melt that produced an easily sheared tribo-interface, as well as the reconstruction of the oxide scale. The sodium carbonate coating with a superior load capability at high temperature is initially reported here. This finding is believed to bring new perspective of high-temperature, extreme-pressure lubrication design with the easily accessible carbonate salt.

With the development of Fully Convolutional Neural Network (FCN), there have been progressive advances in the field of semantic segmentation in recent years. The FCN-based solutions are able to summarize features across training images and generate matching templates for the desired object classes, yet they overlook intra-class difference (ICD) among multiple instances in the same class. In this work, we present a novel fine-to-coarse learning (FCL) procedure, which first guides the network with designed 'finer' sub-class labels, whose decisions are mapped to the original 'coarse' object category through end-to-end learning. A sub-class labeling strategy is designed with unsupervised clustering upon deep convolutional features, and the proposed FCL procedure enables a balance between the fine-scale (i.e. sub-class) and the coarse-scale (i.e. class) knowledge. We conduct extensive experiments on several popular datasets, including PASCAL VOC, Context, Person-Part and NYUDepth-v2 to demonstrate the advantage of learning finer sub-classes and the potential to guide the learning of deep networks with unsupervised clustering.

As one of obvious electrolyte design technologies of Li ion batteries (LIBs) to meet durable high-rate capability at elevated temperatures for battery electric vehicles, this study assesses the superiority of gel polymer electrolyte (GPE) based on experimental results supporting its working mechanism. Our previous study indicated that degradation of cathode electrode under high-rate cycling at elevated temperature was a major cause of the decrease in performance of LIBs, and the single full cells (SFCs) with a GPE designed from dipentaerythritol hexaacrylate and methyl ether methacrylate was re-verified to have superior 3.0C cycling performance at 80 °C. The superiority of the GPE is studied from comparing mid-voltages of discharge profiles of the SFCs, observing the cross-sectional morphology of the electrodes by field emission scanning electron microscopy, assessing the interacting force among the electrolyte components by thermogravimetric analysis, and examining each resistance component of the SFCs by electrochemical impedance spectroscopy. Gel polymerization of liquid electrolyte results in a significant increase of durable high-rate capability of LIBs due to the mechanisms of not only its buffering effect on solvating process of Li+ ions being extracted from the active materials during high rate operation, but also higher thermal stability of electrolyte components, lower susceptibility of the ionic conductivity of the electrolyte to a temperature change, and lower energy barrier to breakup of solvated structures during conducting of Li+ ions in gel polymer matrix.

Grain refinement and homogenization of Cr phase were achieved by laser surface melting (LSM) method, and the properties of Cu-Cr alloy were significantly improved. In this study, LSM of Cu-50Cr alloy (wt.%) was conducted with a high power density (106–107 W/cm2) laser beam, the microstructure and the properties of melt layer were investigated. The size of Cr phase was effectively refined from hundreds of micron scale to several micron scale, and the average size of Cr particles decreased to a few hundred nanometers. High cooling rate effectively inhibited coarsening effect on the Cr particles during liquid phase separation. Spherical Cr particles were dispersed in the melt layer with a thickness of 165 ± 20 μm. Microhardness was obviously enhanced and the maximum hardness was 232HV, which was 2.8 times that of the substrate. Arc duration of the LSM treated contacts increased up to 18%. The withstanding voltage of the fixed and the moving contact increased to 28.7% and 35.4%, respectively. The results show that LSM is an effective method to refine the microstructure of Cu-Cr alloy, and it is a promising modification method for electrical Cu-Cr vacuum contacts.

During high-rate cycling of Li-ion batteries (LIBs) at elevated temperatures, the detachment of the cathode materials from their Al substrate is a major cause of the deterioration in the performance of LIBs. This detachment is suppressed by the addition of an electrolyte additive, poly(ethylene glycol) methyl ether methacrylate, which can act as a buffer zone to prevent the abrupt mass transport of electrolyte within the cathode and as a swing to transport Li+ ions dissociating from the active materials of the cathode. Owing to the dual effects of this type of monomer, an acrylate monomer with one side ether chain, the cathode materials are maintained without detachment from the Al substrate, even under severe cycling conditions. This idea can be applied to LIBs for a series of electric vehicles, which require superior high-rate performance at elevated temperatures.

Semantic segmentation is critical to image content understanding and object localization. Recent development in fully-convolutional neural network (FCN) has enabled accurate pixel-level labeling. One issue in previous works is that the FCN based method does not exploit the object boundary information to delineate segmentation details since the object boundary label is ignored in the network training. To tackle this problem, we introduce a double branch fully convolutional neural network, which separates the learning of the desirable semantic class labeling with mask-level object proposals guided by relabeled boundaries. This network, called object boundary guided FCN (OBG-FCN), is able to integrate the distinct properties of object shape and class features elegantly in a fully convolutional way with a designed masking architecture. We conduct experiments on the PASCAL VOC segmentation benchmark, and show that the end-to-end trainable OBG-FCN system offers great improvement in optimizing the target semantic segmentation quality.

Enterprises are adapting large-scale data processing platforms, such as Hadoop, to gain actionable insights from their "big data". Query optimization is still an open challenge in this environment due to the volume and heterogeneity of data, comprising both structured and un/semi-structured datasets. Moreover, it has become common practice to push business logic close to the data via user-defined functions (UDFs), which are usually opaque to the optimizer, further complicating cost-based optimization. As a result, classical relational query optimization techniques do not fit well in this setting, while at the same time, suboptimal query plans can be disastrous with large datasets. In this paper, we propose new techniques that take into account UDFs and correlations between relations for optimizing queries running on large scale clusters. We introduce "pilot runs", which execute part of the query over a sample of the data to estimate selectivities, and employ a cost-based optimizer that uses these selectivities to choose an initial query plan. Then, we follow a dynamic optimization approach, in which plans evolve as parts of the queries get executed. Our experimental results show that our techniques produce plans that are at least as good as, and up to 2x (4x) better for Jaql (Hive) than, the best hand-written left-deep query plans.