SAS数据挖掘在冷轧酸洗卷质量缺陷分析中的应用

数据挖掘作为一种发现大量数据中潜在信息的数据分析技术,受到各界的密切关注。SAS数据挖掘技术是众多数据挖掘方法中的佼佼者,它在大型企业中得到很好的应用。本文介绍数据挖掘的背景知识,并利用SAS/EM工具,对该技术在冷轧酸洗卷质量缺陷原因分析作了初步尝试。     

LearningADD: Machine learning based acoustic defect detection in factory automation

Defect inspection of glass bottles in the beverage industrial is of significance to prevent unexpected losses caused by the damage of bottles during manufacturing and transporting. The commonly used manual methods suffer from inefficiency, excessive space consumption, and beverage wastes after filling. To replace the manual operations in the pre-filling detection with improved efficiency and reduced costs, this paper proposes a machine learning based Acoustic Defect Detection (LearningADD) system. Moreover, to realize scalable deployment on edge and cloud computing platforms, deployment strategies especially partitioning and allocation of functionalities need to be compared and optimized under realistic constraints such as latency, complexity, and capacity of the platforms. In particular, to distinguish the defects in glass bottles efficiently, the improved Hilbert-Huang transform (HHT) is employed to extend the extracted feature sets, and then Shuffled Frog Leaping Algorithm (SFLA) based feature selection is applied to optimize the feature sets. Five deployment strategies are quantitatively compared to optimize real-time performances based on the constraints measured from a real edge and cloud environment. The LearningADD algorithms are validated by the datasets from a real-life beverage factory, and the F-measure of the system reaches 98.48 %. The proposed deployment strategies are verified by experiments on private cloud platforms, which shows that the Distributed Heavy Edge deployment outperforms other strategies, benefited from the parallel computing and edge computing, where the Defect Detection Time for one bottle is less than 2.061 s in 99 % probability.     

基于图像处理的片式元件缺陷检测算法

表面缺陷检测是电子元器件生产过程中的关键步骤,通过分析表面缺陷检测的现状问题,结合片式元件图像的自身特征,提出一种基于改进的自适应阈值canny算子的片式元件缺陷检测算法;该算法把片式元件灰度图像进行区域分割,得到不同值域范围内的灰度区域,利用带有敏感度的自适应双阈值canny算子进行边缘检测;算法实现结果表明,该算法能高效地检测并处理不同缺陷类型的片式元件,并且可以达到工业生产的检测速度。     

Hesperidin Promotes Osteogenesis and Modulates Collagen Matrix Organization and Mineralization In Vitro and In Vivo

This study evaluated the direct effect of a phytochemical, hesperidin, on pre-osteoblast cell function as well as osteogenesis and collagen matrix quality, as there is little known about hesperidin’s influence in mineralized tissue formation and regeneration. Hesperidin was added to a culture of MC3T3-E1 cells at various concentrations. Cell proliferation, viability, osteogenic gene expression and deposited collagen matrix analyses were performed. Treatment with hesperidin showed significant upregulation of osteogenic markers, particularly with lower doses. Mature and compact collagen fibrils in hesperidin-treated cultures were observed by picrosirius red staining (PSR), although a thinner matrix layer was present for the higher dose of hesperidin compared to osteogenic media alone. Fourier-transform infrared spectroscopy indicated a better mineral-to-matrix ratio and matrix distribution in cultures exposed to hesperidin and confirmed less collagen deposited with the 100-µM dose of hesperidin. In vivo, hesperidin combined with a suboptimal dose of bone morphogenetic protein 2 (BMP2) (dose unable to promote healing of a rat mandible critical-sized bone defect) in a collagenous scaffold promoted a well-controlled (not ectopic) pattern of bone formation as compared to a large dose of BMP2 (previously defined as optimal in healing the critical-sized defect, although of ectopic nature). PSR staining of newly formed bone demonstrated that hesperidin can promote maturation of bone organic matrix. Our findings show, for the first time, that hesperidin has a modulatory role in mineralized tissue formation via not only osteoblast cell differentiation but also matrix organization and matrix-to-mineral ratio and could be a potential adjunct in regenerative bone therapies.     

生产线试运行过程工艺分析软件的应用

介绍了生产线试运行过程中，用工艺分析软件解决所遇到的工艺问题，为工艺分析软件开拓了一个新的应用领域。     

550 MPa级中厚板矫直过程中表面凹坑缺陷的成因与对策

针对550 MPa级高强中厚板在矫直过程中极易出现氧化铁皮压入凹坑缺陷的问题，根据现场矫直工艺，研究了在不同温度下进行热变形时试样表面和横截面的形貌，钢基体与氧化铁皮界面处平直度的变化规律。结果表明，热变形温度为450 ℃时，表面氧化铁皮开始出现裂纹，且随着变形温度的降低，裂纹的数量逐渐增加；此外，氧化铁皮与钢基体之间的界面长度随变形温度的降低而呈直线增加，表明氧化铁皮与钢基体界面平直度随着热变形温度的降低越发粗糙、易脱落而形成凹坑缺陷。为此，应适当提高终轧温度以确保矫直时钢板表面温度大于450 ℃，同时提高钢板的运行速率，防止氧化铁皮在低温变形发生破碎和脱落的同时降低氧化铁皮厚度，从而减小因氧化铁皮脱落而造成凹坑缺陷的发生几率。     

高纯Fe中氢损伤规律的研究

本文采用正电子湮没技术研究了高纯Fe中的氢损伤规律,实验结果表明,在0.5mol/L H_2SO_4溶液中不加毒化剂电解充氢时,产生氢损伤的临界电流密度为20 mA/cm~2;而在0.5mol/L H_2SO_4溶液中加少量As_2O_3作为毒化剂电解充氢时,即使很小的电流密度也会产生氢损伤,配合慢拉伸试验,还发现试样内部产生氢损伤以后,它将控制随后的形变和断裂过程     

Net shape formation of sub-micron alumina with reduced flaws by high-speed centrifugal compaction process

A new compacting method for powder materials is presented. High-Speed Centrifugal Compaction Process (HCP) utilizes centrifugal force of about 10,000 g for the compaction. HCP is suitable for net shape compaction of fine sub-micron and nano-powders. HCP possesses a unique compacting mechanism that is different from other colloidal processes including Pressure Casting (PC), and has a number of useful characteristics. HCP has a higher compacting speed than PC, wide applicability for net shape formation, as well as a defect removing function. Because of homogeneous and flawless green microstructure, HCP alumina shows superior sinterability and higher strength and hardness than most of other aluminas. This article is based on a presentation made in the symposium “The 3rd KIM-JIM Joint Symposium on Advanced Powder Materials“, held at Korea University, Seoul, Korea, October 26–27, 2001 under auspices of The Korean Institute of Metals and Materials and The Japan Institute of Metals.     

涡流检测技术在宝钢热轧轧辊上的应用

介绍了轧辊涡流检测仪的工作要求、设备组成和使用方法,指出对轧辊进行涡流检测可防止爆辊,减少轧辊磨削量。     

冷轧处理线卷取机带头定位精度的改进

针对采用带头与卷取机芯轴橡胶套筒定位区动态接触的传统自动定位技术控制效果不好而带来带头印缺陷的问题，对其产生原因进行了分析，这是因为不同规格、钢种带钢的工艺参数不同，带钢穿带过程中在传送辊上的打滑量不同，在传输通道上的下垂量不同，带头到达卷取机芯轴定位区时计算带头位置与实际带头位置存在不稳定的偏差，因此带头定位精度不稳定。为此，开发了双动态自动定位技术，穿带过程中预留了足够的带钢打滑及下垂富余量，且带头与定位区中心线在助卷皮带咬入口静止状态下接触，消除了影响带头定位精度的主要因素，带头定位精度能稳定在±15 mm的误差范围内。实际应用效果表明，该方法可行、实用。