Sand-bed defect recognition for 3D sand printing based on deep residual network

The 3D sand printing (3DSP), by binder jetting technology for rapid casting, has a pivotal role in promoting the development of the traditional casting industry as a result of producing high-quality and economical sand molds. This work presents an approach for monitoring and analyzing powder sand-bed images to serve as a realtime control system in a 3DSP machine. A deep residual network (ResNet) is used to classify the defects occurring during the powder spreading stage of the process. Firstly, a pre-trained network was applied as the initial parameter; then it was fine-tuned on the labelled defective sample dataset to accomplish the task, which defines the sand-bed defects induced in the 3DSP processing. Furthermore, the recognition and positioning of sand-bed defects were readily achieved by dividing the sand-bed images into blocks. Experiments show that the fine-tuned network has a 98.7% classification accuracy on the validation dataset of sand-bed defects and 95.4% recognition accuracy for the sand-bed images.     

Additives to prevent the formation of surface defects during poly(vinyl chloride) calendering

Gas checks are visible fleck-shaped defects that occur on the surface of poly(vinyl chloride) (PVC) films during industrial calendering. Films containing these surface defects often do not meet minimum product specifications and therefore must be disposed of or recycled, resulting in increased cost and material waste. Currently, gas checks are controlled by keeping film gauge low and through trial-and-error modifications of processing parameters by calender operators. In this work, our group developed a series of chemical additives that can be blended with PVC to prevent the formation of gas check defects. We found that a series of poly(caprolactone) (PCL)-based compounds with diester linkers and alkyl chain cappers were all effective at preventing the formation of gas checks during calendering, with additive concentrations as low as 8 phr producing films with no gas checks. We found that the blends produced with our additives had higher melt viscosities than those produced with additives that do not remove gas checks, suggesting that viscosity plays an important role in preventing gas check defects.     

Excitation with phase shifted fields-enhancing evaluation of deep cracks in eddy-current testing

This study enhances the applicability of eddy current testing to the evaluation of deep surface-breaking cracks in the front face of thick structures, while keeping the advantages of higher frequency inspection. The key idea is to suppress eddy currents on the surface of a test-piece, and thus realize the deeper penetration of eddy currents. Based upon this idea, a mutual induction eddy current testing probe, consisting of four coaxial rectangular tangential exciter coils and a pancake pick-up coil, has been designed. The exciters are driven by AC currents with different phases and amplitudes, which contrive to emulate the desired behavior. Numerical simulations and consecutive experimental verifications are presented in the paper. It is revealed that when the new probe is used for inspections under a frequency of 10 kHz there is a 20% difference between signals due to an electro-discharge machined (EDM) notch of 15 mm depth and those due to one of 20 mm depth. On the other hand, a conventional probe driven with the same frequency gives only 1% difference between the signals of the two EDM notches.     

Minimization of defects in nickel-based superalloy weld overlay deposited by the GTAW cold wire feed process

The objective of the present work was to evaluate the influence of the welding parameters in the formation of defects in weld overlays with nickel-based alloys, and its possible elimination through the correct adjustment of the welding parameters. Coatings were deposited with the nickel-based alloy types Inconel 625, Hastelloy C276 and Inconel 686 on C–Mn steel plates, using GTAW cold wire feed process. The planning of the experiments was accomplished by applying the Taguchi method. The control factors evaluated were the energy technique (ET), the welding heat input (E), the type of alloy (L), the shielding gas (G) and the type of arc oscillation (T). Other parameters were maintained constant, having previously been investigated. The results showed that the arc oscillation type is spiral; although it contributes significantly to reduce the dilution level, it causes a strong instability to the process, resulting in most of the cases in surface defects or defects among passes. The optimum condition to avoid defects among passes was identified by the Taguchi method, which was constituted by the following combination of control factors: 2-2-2-3-3, or by: I energy technique (TE-I); medium heat input level (E_medium); Hastelloy C276 alloy; shielding gas Ar + He; double-8 arc oscillation. The optimum condition for the welding without defects results in high dilution level not being indicated for the welding of resistant coverings to the corrosion.     

机器人喷涂缺陷分析与改进

以FANUC P200E型喷涂机器人为例,分析了机器人喷涂过程中产生的缺陷问题,提出了程序优化、设备改进和维护措施。     

PE木器底漆常见问题及对策

针对PE底漆在运用过程中出现的弊病,探讨了PE底漆的冬季打磨、调油防沉、加白水后消泡和填充性能。另外将着重对PE清底漆的抗下陷性进行探讨,并给出了相应的解决方案。     

纤维-金属层板粘接质量的非线性超声评价研究

传统超声检测利用超声波传播中遇到缺陷时波的反射、散射等特征进行缺陷检测和评价,实际上表征的是缺陷和周围介质的声阻抗差别。当超声波在材料中传播时,位错等微小缺陷会与其发生非线性相互作用,非线性超声检测就是利用这些非线性响应信号进行材料性能的评估和微小缺陷的检测,本质上反映的是微小缺陷对材料非线性的影响。文中从非线性超声理论出发,提出运用2阶和3阶非线性系数进行纤维-金属粘接层的评价,实验结果表明非线性系数可以表征粘接缺陷。     

ZLW10型矫直机的设计及应用

介绍了一种矫直机,用于U型梁在辊压生产过程中出现侧弯、翘曲、扭曲等缺陷时的矫直。该矫直机具有精确的量化调节能力,可实现针对U型梁的自动矫形功能。     

铸造条件对重力铸造TiAlNb铸件缩孔及夹杂气泡的影响

研究了感应凝壳炉熔炼的TiAlNb合金的陶瓷型壳重力浇铸。使用了扫描电镜（SEM）,能谱仪（EDX）以及X射线光电子能谱分析（XPS）对铸造合金的显微组织及气孔缺陷进行了检测分析。研究发现,氧化膜在气孔及缩孔中存在并且是这些缺陷的不均匀形核位置,而这些气孔缺陷一方面是浇铸过程中气体夹带形成的,另外一方面是气体小分子积聚形成。这些气孔广泛存在于铸件的表层及中心部分。铸件直径,模具温度,以及氩分压都对气孔缺陷的形成有一定的影响。     

基于优化Gabor滤波器的铸坏表面缺陷检测应用研究

目的提高金属铸坯表面缺陷检测精度。方法由于金属铸坯表面上存在鱼鳞状构造,其亮度和背景区域纹理特征不一致,而且有缺陷和无缺陷的区域的灰度值极其相似,使得缺陷非常难以准确检测出来。为解决上述问题,以便更有效地检测表面缺陷,通过详细分析金属铸坯表面缺陷特征,将该类零件表面缺陷分为两种类型,提出一种基于优化Gabor滤波器的金属表面缺陷检测算法,该算法通过设计两种评价函数,利用评价函数最大限度地提高无缺陷和缺陷区域之间的能量差,以选取Gabor滤波器四个最佳参数,同时使用双阈值滤波方法,以减少由于噪声和伪缺陷引起的测量误差。结果利用3种滤波算法对四十幅带有缺陷的图像进行试验,实验表明该算法在角部裂纹、细裂纹和伪裂纹检测精度分别达到92.50%、92.50%和95.50%。结论 Opt-Gabor算法能根据已分类的两种不同类型的裂纹较为准确地检测出铸坏表面缺陷,在测量精度上略优于其他几种算法。