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。     

Electrocatalytic Reduction of CO2 to Ethylene by Molecular Cu‐Complex Immobilized on Graphitized Mesoporous Carbon

The electrochemical reduction of carbon dioxide (CO₂) to hydrocarbons is a challenging task because of the issues in controlling the efficiency and selectivity of the products. Among the various transition metals, copper has attracted attention as it yields more reduced and C2 products even while using mononuclear copper center as catalysts. In addition, it is found that reversible formation of copper nanoparticle acts as the real catalytically active site for the conversion of CO₂ to reduced products. Here, it is demonstrated that the dinuclear molecular copper complex immobilized over graphitized mesoporous carbon can act as catalysts for the conversion of CO₂ to hydrocarbons (methane and ethylene) up to 60%. Interestingly, high selectivity toward C2 product (40% faradaic efficiency) is achieved by a molecular complex based hybrid material from CO₂ in 0.1 m KCl. In addition, the role of local pH, porous structure, and carbon support in limiting the mass transport to achieve the highly reduced products is demonstrated. Although the spectroscopic analysis of the catalysts exhibits molecular nature of the complex after 2 h bulk electrolysis, morphological study reveals that the newly generated copper cluster is the real active site during the catalytic reactions.     

A facile method for preparation of PGS@ZB‐N and gas‐sol alternating synergistic effect for fire resistance of EVA

Through the simple precipitation of palygorskite (PGS) by zinc borate (ZB) (to make PGS@ZB) and the decoration of PGS@ZB by dodecylamine (N), a novel organic‐inorganic@inorganic hybrid flame retardant of PGS@ZB‐N was prepared and was incorporated with ethylene vinyl acetate copolymer (EVA) to improve its flame retardance. The structure and morphology of PGS@ZB‐N were characterized by Fourier transform infrared (FTIR) spectroscopy, X‐ray diffraction (XRD), and scanning electron microscopy (SEM), and it was confirmed that the PGS@ZB‐N hybrid had been successfully prepared. The flame retardancy and burning behavior of EVA/PGS@ZB‐N/EG (EG = expandable graphite) composite were studied through thermogravimetric analysis (TGA), limiting oxygen index (LOI), UL‐94 (by the vertical burning test), and cone calorimeter test (CCT) characterizations. The prepared EVA/PGS@ZB‐N/EG composite obtained an LOI value of 41.2% with the addition of 30 wt% PGS@ZB‐N/EG. It was found that EVA/PGS@ZB‐N/EG was protected through a gas phase and condensed phase alternating synergistic effect mechanism.     

Effect of fortification with asparagus powder on the qualitative properties of processed cheese

The qualitative properties of processed cheese (PC) fortified with different levels of asparagus powder (AP) (0.5%, 1% and 1.5% wt/wt) were evaluated during storage. AP decreased the pH and lipolysis indexes and increased the phenolic content, antioxidant activity and proteolysis of the processed cheeses. AP made the structure of the cheese more elastic, increased the rigidity and decreased the spreadability compared with the control sample, which corresponded to the results obtained using dynamic oscillatory rheometry. The results showed that AP as a rich source of bioactive components could be used for the fortification of processed cheeses.     

Random poly(ε‐caprolactone‐L‐alanine) by direct melt copolymerization

A series of random polyesteramides (PEAs) with a range of molar composition from 90/10 to 50/50 were synthesized by direct melt polycondensation of ε‐caprolactone and l ‐alanine. Their structure was fully characterized by Fourier transform IR and NMR spectroscopy. The resulting copolymers are completely amorphous with the exception of PEA‐90/10 which possesses a semicrystalline structure. These PEAs present increasing glass transition temperatures at increasing l ‐alanine contents and exhibit fairly good thermal stability with 10% mass loss temperatures reaching 315 °C. © 2020 Society of Industrial Chemistry     

基于特征融合的中文简历解析方法研究

针对基于规则和统计的传统中文简历解析方法效率低、成本高、泛化能力差的缺点，提出一种基于特征融合的中文简历解析方法，即级联Word2Vec生成的词向量和用BLSTM（Bidirectional Long Short-Term Memory）建模字序列生成的词向量，然后再结合BLSTM和CRF（Conditional Random Fields）对中文简历进行解析（BLSTM-CRF）。为了提高中文简历解析的效率，级联包含字序列信息的词向量和用Word2Vec生成的词向量，融合成一个新的词向量表示；再由BLSTM强大的学习能力融合词的上下文信息，输出所有可能标签序列的分值给CRF层；再由CRF引入标签之间约束关系求解最优序列。利用梯度下降算法训练神经网络，使用预先训练的词向量和Dropout优化神经网络，最终完成对中文简历的解析工作。实验结果表明，所提的特征融合方法优于传统的简历解析方法。