光伏用电致发光缺陷检测仪空间分辨率的量化评估

针对光伏用电致发光缺陷检测仪空间分辨率目视判定重复性、准确性差,并且难以量化等问题,提出一套基于卷积神经网络模型的空间分辨率量化评估方法。参照相关标准JJF(闽)1088-2018,采取不同的拍摄条件拍摄并切割出空间分辨率线对图像,对图像进行人工分类。设计卷积神经网络结构,采用卷积层、池化层和全连接层结构,并使用已分类完成的空间分辨率线对图像对模型进行训练。最终使用测试集对模型进行评估,结果表明,模型在测试集上的判别正确率达到99.2%。该方法满足使用要求,可取代目视判别,提高了光伏用电致发光缺陷检测仪空间分辨率判定的准确性与重复性,并为其量化提出了新的解决思路。由于太阳电池片用光致发光缺陷检测仪与光伏用电致发光缺陷检测仪的成像方式类似,该方法可兼容太阳电池片光致发光检测仪的性能评估。     

Smart optical measurement probe for autonomously detecting nano-defects on bare semiconductor wafer surface: Verification of proposed concept

We propose a new method of inspecting a surface for fine defects that combines the optical inspection method with observation of the physical behavior of a liquid. A liquid thin film on a substrate behaves as a near-field physical probe that autonomously captures nano-particulate defects. Optical observation of the interfacial behavior of the liquid thin film is used to detect minute defects. This method combines the characteristics of optical detection (i.e., detection from a remote field and simultaneous detectability on a plane) and the high sensitivity of a physical near-field probe. We examined the basic principles of the proposed method through numerical calculation and applied it in experiments to detect fine particulate defects on a silicon substrate for semiconductor manufacturing to demonstrate the validity of the basic concept of the proposed method.     

综采工作面端面破碎顶板控制技术应用

针对8103综采工作面过断层应力区期间出现端面距加大,端面破碎现象严重,降低了工作面回采效率,影响支架支护效果,局部区域出现冒落现象,北辛窑矿通过技术研究,对过断层应力区端面破碎顶板维护方案进行了探讨,决定采用注浆锚杆支护技术。通过实际应用,控制了端面顶板破碎现象,降低了端面距,保证了工作面安全高效回采。     

基于BIM技术的煤矿井下支护优化

BIM技术作为一种信息模式化的手段,被广泛运用于各类生产实践。为探讨BIM技术在矿井支护中的应用方式,在论述了矿井综采工作面巷道支护优化设计的基础和可能性后,详述了新型锚索的具体优化方案和优化后的优点,并以新型锚索支护技术为例开展了具体分析。在工艺改革和支护材料的管理及选择方面,提出了推引锚固技术、恒阻大变形锚索支护以及严格支护材料质量检验制度的建议。结合该技术进行支护方案的优化设计,具有一定可行性,且更加经济合理。     

焊缝表面缺陷激光扫描三维重构测量

为检测工件焊缝表面缺陷，采用点激光位移传感测距法和数据拟合技术进行焊缝表面缺陷检测试验研究. 首先采集缺陷表面轮廓数据点，利用高斯滤波对原始数据降噪处理. 再对处理后的数据点进行Delaunay三角剖分，使散乱点连接并结合其空间坐标重构出缺陷的三维模型. 结果表明，基于点激光位移传感测距技术及焊缝表面缺陷三维重构方法可以准确判断焊缝成形情况.     

Preparation of molecularly imprinted fluorescence sensor based on carbon quantum dots via precipitation polymerization for fluorescence detection of tetracycline

A facile and effective method was proposed to prepare the molecularly imprinted fluorescence sensor with carbon quantum dots, which were modified vinyl groups by acrylic acid on the surface. The obtained fluorescence composite material was investigated by transmission electron microscope and Fourier transform infrared spectra. After the experimental conditions were optimized, a linear range of 1.0–60 μmol L⁻¹ was obtained and the detection limit was 0.17 μmol L⁻¹. The novel fluorescence sensor can be successfully used to detect tetracycline in real samples. This study provides a convenient strategy for selective recognition and rapid detection of tetracycline in the complex environment.     

深度学习识别光网络单元故障的设计与应用

为解决依赖装维上门鉴别光网络单元故障带来的不便,可以从机器视觉入手实现自动化故障识别。近年,ImageNet挑战赛的成功推动了物体识别技术的跨越式发展,特别是基于卷积的深度学习技术在视觉识别方面已经达到人类水平,为光网络单元故障的自动识别提供了技术基础。文章对识别光网络单元的工作状态进行了研究,将设备工作状态分为7个场景,提出了利用手机APP采集图片识别故障的解决方案并投入了实际生产;重点阐述了深度学习模块的设计与实现,提出一种通过算法整合的方式综合运用物体检测和图像分类算法,分3阶段逐步求精,解决了图片过滤,光网络单元型号和状态识别等问题,实现了基于计算机视觉自动识别光网络单元故障。从数据上看产品的端到端准确率超过84%,识别速度达到10 FPS,月均提供服务超过1万人次,在减少用户等待的同时节约了人力资源。     

Estimating posterior inference quality of the relational infinite latent feature model for overlapping community detection

Overlapping community detection has become a very hot research topic in recent decades, and a plethora of methods have been proposed. But, a common challenge in many existing overlapping community detection approaches is that the number of communities K must be predefinedmanually. We propose a flexible nonparametric Bayesian generative model for count-value networks, which can allow K to increase as more and more data are encountered instead of to be fixed in advance. The Indian buffet process was used to model the community assignment matrix Z, and an uncollapsed Gibbs sampler has been derived.However, as the community assignment matrix Z is a structured multi-variable parameter, how to summarize the posterior inference results and estimate the inference quality about Z, is still a considerable challenge in the literature. In this paper, a graph convolutional neural network based graph classifier was utilized to help to summarize the results and to estimate the inference quality about Z. We conduct extensive experiments on synthetic data and real data, and find that empirically, the traditional posterior summarization strategy is reliable.     

后背门内板密封面起皱问题的改善

对后背门内板拉伸过程出现的起皱问题进行了分析,并结合CAE分析制定整改方案,通过现场模具维修,消除了该处起皱问题,对类似制件的开发和模具维修有一定的借鉴作用。     

上榆泉煤矿煤泥水多参数在线检测系统应用研究

针对上榆泉煤矿煤泥水性质参数人工检测周期长、误差大、滞后性严重的问题,采用多参数传感器和PLC,开发了煤泥水多参数在线检测系统。经过8 h稳定运行,平均结果显示该厂煤泥水参数为pH值为7.71、浓度9.45%、中值直径(D50)39.56μm、流量2 440 m3/h、水质硬度428.94 mg/L,与人工检测结果相一致。应用结果表明,该检测系统能够实现煤泥水多参数在线监测和数据分析功能,使煤泥水的处理更高效。