基于改进YOLOv4的扫描电镜磨粒图像智能识别

磨损颗粒分析是设备磨损故障诊断和预测的有效手段，为了提高磨粒检测的自动化和智能化程度，提出1种基于改进YOLOv4的目标检测算法，并应用于航空发动机扫描电镜磨粒图像识别.首先，新算法采用VoVNetv2-39替换YOLOv4原主干网络CSPDarknet53，并引入BiFPN特征金字塔结构与新主干相连，同时调整模型中所有3×3标准卷积为深度可分离卷积，以加强多层次特征融合，构造轻量级网络；其次，利用迁移学习解决扫描电镜磨粒图像数量较少的问题，并通过冻结训练加速模型训练过程；最后，应用实际发动机扫描电镜磨粒图像验证，结果表明：新算法相较于原YOLOv4网络，在保证精度的前提下，网络参数量大幅降低，识别速度提升51.1%，满足实际扫描电镜磨粒图像快速、简洁和高精度的检测需求，具备潜在的工程应用价值.     

基于背景色彩识别的磨粒图像分割方法

针对铁谱图像背景色彩相对单一的特点,设计了1种基于背景色彩识别的磨粒图像分割方法,采用该方法对轴尖-宝石轴承的润滑油铁谱图像进行图像分割,并用所得磨粒覆盖面积推算轴尖磨损量.通过大量磨粒图像分割实践表明,所设计的方法具有精度高、计算速度快及适应性强等优点.定量磨损分析表明,该摩擦副的磨损进程极为缓慢.     

卷积神经网络在装备磨损颗粒识别中的研究综述

铁谱法是用于装备故障诊断的1种重要方法,其中铁谱法的重点是铁谱图像的分析,即磨损磨粒分析. 卷积神经网络是当下最流行的深度学习算法之一,其广泛应用于图像识别领域,使得图像识别领域得到突破. 随着卷积神经网络的快速发展,磨损颗粒在智能识别方面的技术取得了重大的突破. 本文中首先简述了卷积神经网络与磨粒智能识别的发展历史,针对基于卷积神经网络的磨粒识别方法进行了从图像数据集处理到模型优化技术方面的介绍,并详细说明了这些技术在磨粒识别中的具体应用实例. 然后从现有网络和自设计网络两方面分类,整理了近年来卷积神经网络应用于磨粒智能识别的代表性文献,综述了这些工作所提出的模型结构和特点,分析并阐述了各个模型主要的识别原理,各个网络结构存在的优缺点,以及它们的数据采用情况等,并对未来磨粒智能识别的主要研究方向进行了展望. 最后肯定了卷积神经网络方法在磨粒智能识别方面的重要性,同时指出了基于此方法的磨粒识别模型的缺点,并提出了应紧跟图像识别领域的最新技术以促进磨粒智能识别水平提高等建议,对磨粒智能识别的发展具有一定的意义.      

因子模糊化BP神经网络在磨粒识别中的应用

在引入磨粒形态学描述了提取磨损颗粒显微形态特征的基础上,用人工神经网络技术,编制了用于磨损颗粒自动识别的ＢＰ网络计算机模拟程序,应用所引入的因子模糊化训练法可使训练速度加快,以异或问题为例,速度可提高５～１０倍。用此网络对磨粒测试库进行识别实验发现,识别速度快且正常率在９０％以上,优于传统的磨粒识别方法。     

基于粗集—神经网络的磨粒模式识别

应用粗集理论中最小约简的近似算法 ,对磨粒的形状参数进行约简 ,找出判断磨粒模式的形状参数长短轴比 Rt和圆度 Rd,同时采用这些参数训练神经网络以进行磨粒模式识别 .结果表明 :应用此算法 ,对 63个已知样本和历时 2年多对柴油机 1 4 5个润滑油油样进行制谱分析判断 ,准确率在 90 %以上 ,比原来用模糊识别的准确率提高了约 1 0个百分点 ;使用 BP网络减化了网络结构 ,使网络的训练速度加快 ,整个系统变得简单、可靠、有效     

基于径向基函数神经网络的磨粒识别系统

应用磨粒形状特征参数、颜色特征参数和表面纹理特征参数对磨粒形态进行量化表征，并以此为输入矢量，引入径向基函数神经网络对磨损微粒进行自动分类识别，建立了适用于磨粒识别的径向基函数神经网络模型，并给出了具体算法．应用实例表明，径向基函数神经网络的收敛速度和识别率优于传统的BP神经网络．     

基于图像的磨粒在线监测系统关键技术研究

针对磨粒在线监测系统开发要求,对其中的关键技术进行了深入研究.构建了基于显微图像分析和微泵进样的磨粒在线监测系统,给出了图像检测子系统和微流控分析芯片的详细设计原理和方法;利用粗糙集理论对磨粒特征参数进行了优化;根据系统光路特点,对磨粒图像进行了基于彩色特征的转换,并通过与背景图像的差值处理来快速提取磨粒目标.最后,利用颗粒计数器和铁谱分析技术对系统性能进行了检验,结果表明本系统磨粒统计与识别准确率分别达到90%和95%,具有较高的检测精度,满足磨粒在线监测要求.     

磨损数值仿真技术的研究进展

综述了近年来国内外磨损数值仿真技术研究进展和现状，重点评述了数值仿真方法在磨粒磨损、疲劳磨损、磨损表面形貌模拟和实际摩擦副磨损研究方面的应用．指出磨粒磨损仿真主要集中于二体磨损的磨损表面形貌分析和磨损预测;疲劳磨损仿真主要集中基于断裂力学方法的建模以及基于有限元方法的参数定量计算和疲劳裂纹行为数值模拟；实际摩擦副磨损仿真主要侧重于齿轮、凸轮和缸套-活塞环等典型摩擦副的磨损研究．就当前磨损数值仿真研究的主要问题和发展方向提出了建议。     

一种新的磨粒图像特征参数

探讨了小波分析在磨粒图像处理中的应用 ,提出了基于磨粒图像统计特征的粒度参数概念 .利用小波分解对磨粒图像进行分解 ,然后对分解的图像进行统计分析 ,实现磨粒图像的特征提取 .结果表明 ,粒度参数可以作为磨粒图像的综合特征参数     

含磨粒油润滑齿轮胶合失效的计算模型与试验研究

在机械设备的运转过程中，齿轮润滑油都含有很多不同尺寸的磨粒，研究表明，磨粒产生的机械效应和热效应都不容忽视，前者可以导致齿轮的磨粒磨损，后者能够造成齿轮的胶合失效。尽管目前人们已经对磨粒导致胶合失效的机理有所认识，但还没有建立起相应的计算方法和模型，有关的试验分析方法也还没有看到文献报道。因此，从机理，计算方法和试验等方面对含磨粒润滑工况下的齿轮胶合失效进行了分析与研究，在给定的条件下，通过几种含     

Deep learning-based on-line image analysis for continuous industrial crystallization processes

In situ microscopic imaging is a useful tool in monitoring crystallization processes, including crystal nucleation, growth, aggregation and breakage, as well as possible polymorphic transition. To convert the qualitative information to be quantitative for the purpose of process optimization and control, accurate analysis of crystal images is essential. However, the accuracy of image segmentation with traditional methods is largely affected by many factors, including solid concentration and image quality. In this study, the deep learning technique using mask region-based convolutional neural network (Mask R-CNN) is investigated for the analysis of on-line images from an industrial crystallizer of 10 m³ operated in continuous mode with high solid concentration and overlapped particles. With detailed label points for each crystal and transfer learning technique, two models trained with 70,908 and 7,709 crystals respectively are compared for the effect of training data amount. The former model effectively segments the aggregated and overlapped crystals even at high solid concentrations. Moreover, it performs much better than the latter one and traditional multi-scale method both in terms of precision and recall, revealing the importance of large number of crystals in deep learning. Some geometrical characteristics of segmented crystals are also analyzed, involving equivalent diameter, circularity, and aspect ratio.     

水润滑条件下磨粒尺寸对橡胶密封副摩擦学行为的影响

磨粒磨损是各类机械橡塑密封装备的一种典型失效形式. 本文中开展了水润滑条件下磨粒尺寸对O型橡胶密封配副摩擦学行为的影响研究，分析了不同颗粒尺寸下的摩擦系数时变特性，探讨了不同运行阶段下的磨损形貌、颗粒运动特性和损伤失效机制. 结果表明:颗粒尺寸对橡胶/金属密封副的摩擦学性能有重要影响；研究发现了影响摩擦系数和损伤机制的两个颗粒尺寸临界值；当颗粒尺寸大于临界值约75 μm时，颗粒难以穿过密封界面，但少数颗粒能嵌入到摩擦副的接触区两侧，犁削作用下的金属表面产生较深的犁沟；当颗粒尺寸介于两临界值内时，接触副两侧形成“颗粒嵌入带”，颗粒以单独或颗粒群形式嵌入到橡胶内，尽管犁削了配副金属但也起到了良好的承载作用，表现出较低的摩擦系数并减缓了橡胶的磨损；当颗粒尺寸约小于另一临界值12.5 μm时，颗粒能较自由地通过摩擦界面，对配副金属表面起到了抛光效应，但也加速了橡胶的冲蚀磨损，橡胶磨损表面呈现“突脊-犁沟-突脊”交替特征，因此工程上应尽量避免此类现象的发生.      

流体润滑状态下磨粒尺寸对液压泵磨损的影响

本文研究了流体润滑状态下磨粒尺寸对液压泵磨损的影响。结果表明,在液压系统中,泵的磨损主要决定于磨粒与泵内关键运动副动态间隙之间的尺寸关系:尺寸小于动态间隙的磨粒影响很小;尺寸明显比动态间隙大的磨粒不会对泵产生磨损作用;尺寸与动态间隙很接近的磨粒能使泵发生严重磨损。作者还利用扫描电子显微镜对流体润滑状态下三体磨粒磨损中的磨屑形貌进行了观察与分析,并且提出了控制液压系统中磨粒损磨的可行施措。     

Rod-like particles matching algorithm based on SOM neural network in dispersed two-phase flow measurements

A matching algorithm based on self-organizing map (SOM) neural network is proposed for tracking rod-like particles in 2D optical measurements of dispersed two-phase flows. It is verified by both synthetic images of elongated particles mimicking 2D suspension flows and direct numerical simulations-based results of prolate particles dispersed in a turbulent channel flow. Furthermore, the potential benefit of this algorithm is evaluated by applying it to the experimental data of rod-like fibers tracking in wall turbulence. The study of the behavior of elongated particles suspended in turbulent flows has a practical importance and covers a wide range of applications in engineering and science. In experimental approach, particle tracking velocimetry of the dispersed phase has a key role together with particle image velocimetry of the carrier phase to obtain the velocities of both phases. The essential parts of particle tracking are to identify and match corresponding particles correctly in consecutive images. The present study is focused on the development of an algorithm for pairing non-spherical particles that have one major symmetry axis. The novel idea in the algorithm is to take the orientation of the particles into account for matching in addition to their positions. The method used is based on the SOM neural network that finds the most likely matching link in images on the basis of feature extraction and clustering. The fundamental concept is finding corresponding particles in the images with the nearest characteristics: position and orientation. The most effective aspect of this two-frame matching algorithm is that it does not require any preliminary knowledge of neither the flow field nor the particle behavior. Furthermore, using one additional characteristic of the non-spherical particles, namely their orientation, in addition to its coordinate vector, the pairing is improved both for more reliable matching at higher concentrations of dispersed particles and for higher robustness against loss of particle pairs between image frames.     

水介质中丁腈橡胶溶胀机理及其对磨粒磨损行为影响

研究首先对不同丙烯腈质量分数的丁腈橡胶（18%、26%和41%,即N18、N26和N41）进行浸泡试验分析基于溶胀时间的静态溶胀行为. 然后进一步对溶胀前后的丁腈橡胶样品进行圆型及角型磨粒单向滑动磨粒磨损试验. 为了揭示丁腈橡胶的溶胀机理及溶胀对不同粒型作用下磨粒磨损行为的影响,对试样进行质量、硬度、交联链比重（NMR）和表面形貌（SEM）测试. 结果表明:丁腈橡胶溶胀度随时间的增加而增加,并于240 h后趋于平衡,溶胀影响下老化的主要原因为交联链破坏和物质溶解.丁腈橡胶硬度随溶胀时间的增加而降低. 溶胀后摩擦系数和磨损量增加,同粒径下磨粒磨损程度角型大于圆型,角型磨损受溶胀影响程度更大,其最大溶胀后磨损增量为圆型磨粒的1.56倍. 圆型和角型磨粒分别以滚动和滑动的方式进入摩擦副,形成压入坑和嵌入坑两种不同的表面形貌,溶胀后橡胶组织疏松软化导致表面凹坑加深. 丁腈橡胶溶胀及其对磨粒磨损的影响程度随丙烯腈含量的增加而减小.      

A slip-line plasticity analysis of abrasive wear of a smooth and soft surface sliding against a rough (fractal) and hard surface

Abrasive wear of a soft and smooth surface sliding against a rough (fractal) and hard surface was analyzed by the slip-line theory of plasticity. The analysis is based on a slip-line model of a rigid spherical asperity (wear particle) plowing through a soft surface and removing material by microcutting. Integration of this single-contact model into a three-dimensional contact mechanics analysis of an abrasive surface exhibiting multi-scale roughness described by fractal geometry yielded relationships of the abrasive wear rate and wear coefficient in terms of the interfacial shear strength (adhesion), topography (fractal) parameters of the hard/rough surface, elastic–plastic material properties of the soft/smooth surface, and total normal load. Analytical results from the single-contact analysis provide insight into the deformation of a perfectly plastic material caused by the abrasive action of a rigid asperity/wear particle under different normal load and interfacial friction (adhesion) conditions. The dependence of the abrasive wear rate and wear coefficient on normal load (global interference), roughness of the abrasive surface, elastic–plastic material properties of the abraded surface, and interfacial shear strength (lubrication effect) is interpreted in the context of numerical results obtained for representative ceramic/ceramic, ceramic/metallic, and metal/metal sliding systems.