基于改进的Random Subspace的客户投诉分类方法

电信业的客户投诉不断增多而又亟待高效处理。针对电信客户投诉数据的特点，提出了一种面向高维数据的改进的集成学习分类方法。该方法综合考虑客户投诉中的文本信息及客户通讯状态信息，基于Random Subspace方法，以支持向量机（Support Vector Machine，SVM）为基分类器，采用证据推理（Evidential Reasoning，ER）规则为一种新的集成策略，构造分类模型对电信客户投诉进行分类。所提模型和方法在某电信公司客户投诉数据上进行了验证，实验结果显示该方法能够显著提高客户投诉分类的准确率和投诉处理效率。     

适用于无线体域网的动态口令认证协议

无线体域网中传输的是与生命高度相关的敏感数据，身份认证是信息安全保护的第一道防线。现有的基于人体生物信息的身份认证方案存在信息难提取、偶然性大和误差性大的问题，基于传统密码学的认证方案需较大计算资源和能量消耗，并不适用于无线体域网环境。为此，在动态口令和非对称加密机制基础上，提出一种适用于无线体域网的动态口令双向认证轻量协议，并对其进行形式化分析。通过理论证明、SVO逻辑推理及SPIN模型检测得出：该协议满足双向认证，且能够抵御重放攻击、伪装攻击、拒绝服务器攻击和口令离线攻击，具有较高安全性。     

基于碳纳米管的超级电容器聚吡咯电极复合方法

For the defect that the mechanical properties of polypyrrole supercapacitors decrease with charge and discharge cycles, and the cycle stability is poor, a compound method of polypyrrole electrode for supercapacitors based on multi-walled carbon nanotubes is proposed. The structure of the formed polypyrrole coated multi-walled carbon nanotubes effectively increase the specific surface area of the electrode material, the utilization rate of the active material and the electrical conductivity, improve the rapid charge and discharge performance of the electrode material, and greatly improve cyclic stability of polypyrrole. The composite electrode materi- al of polypyrrole and functionalized multi-walled carbon nanotubes for supercapacitors is prepared by pulse current deposition meth- od. It is scanned at a scanning rate of 1000mV?s -1 in a 3mol?L -1 potassium chloride electrolyte, after 100,000-cycle, the capacity on- ly declines by 16%.     

矿用直线振动筛结构故障分析

针对煤矿选煤厂直线振动筛工作过程中出现的结构故障问题,采用有限元仿真与现场测试结合的方法对振动筛出现故障的原因进行了分析研究。介绍了振动筛的主要结构及常见故障,并以某矿选煤厂发生大梁断裂的振动筛为例,对该振动筛进行三维建模和有限元模态分析,分析结果表明:随着模态阶次的增加,振型的变化逐渐由移动转为结构本身的扭曲变形。在该振动筛空载工作状态下,选用本质安全型便携式测振记录仪进行振动测试并处理测试信号,得到该振动筛的工作频率。结合模态分析结果和实际测试结果,判断出空载时频繁运行振动筛是造成其大梁断裂的主要原因。根据故障原因给出维护建议:应当尽量避免振动筛在空载或物料较少的状态下长时间运行;在振动筛投入使用前,应预先了解振动筛的工作状态及可能存在的薄弱环节,尽可能避免结构故障,保证振动筛可靠运行。     

Isolation and identification of two major acylated anthocyanins from purple sweet potato (Ipomoea batatas L. cultivar Eshu No. 8) by UPLC‐QTOF‐MS/MS and NMR

In recent years, researches on the isolation and preparation of monomeric anthocyanins have intensified because of the requirements of quantitative and structure–bioactivity relationship analyses. However, simple and effective methods about the scale of monomeric anthocyanins from the natural purple sweet potato powder are rarely reported. In this study, high molecular weight acylated monomeric anthocyanins were isolated from purple sweet potato (Ipomoea batatas L. cultivar Eshu No. 8) via the combination of column chromatography and semi‐preparative HPLC technology and identified mainly by ultra‐high‐performance liquid chromatography quadrupole time‐of‐flight tandem mass spectrometry/mass spectrometry (UPLC‐QTOF‐MS/MS) and ¹H and ¹³C nuclear magnetic resonance (NMR). Two major acylated anthocyanins were unambiguously determined as peonidin 3‐O‐(6‐O‐(E)‐caffeoyl‐(2‐O‐(6‐O‐p‐hydroxybenzoyl)‐β‐D‐glucopyranosyl)‐β‐D‐glucopyranoside)‐5‐O‐(β‐D‐glucopyranoside) and peonidin 3‐O‐(6‐O‐(E)‐caffeoyl‐(2‐O‐(6‐O‐(E)‐feruloyl)‐β‐D‐ glucopyranosyl)‐β‐D‐glucopyranoside)‐5‐O‐(β‐D‐glucopyranoside). The results of this study may help promote the purification of high molecular weight acylated anthocyanins from purple sweet potato as well as from other plant materials in nature.     

基于气动性能与刚度特性的风力机翼型优化设计

结合翼型泛函集成理论与叶片截面刚度矩阵数学计算模型,提出了风力机中等厚度翼型气动性能与结构刚度特性的一体化设计方法,实现了翼型气动性能与叶片截面刚度特性的同时提高。对考虑叶片截面铺层参数变化设计的WQ-B300翼型与DU97-W-300翼型进行了气动性能与结构刚度特性对比分析,结果表明:相比于DU97-W-300翼型,WQ-B300翼型的气动性能与叶片截面刚度性能均有显著提高,其挥舞刚度和摆振刚度分别提高了6.2%和8.4%,验证了该设计方法的可行性,给风力机中等厚度及大厚度翼型设计提供了一种思路。     

基于RGV轨导柔性线环境离散型零件生产过程管理与实践

为满足航空中小型结构件快速交付的需要,以提升航空中小型结构件制造高精度、自动化和高效率为目标,通过对航空中小型结构件加工现状进行梳理分析,搭建柔性生产线自动排程与调度平台,实现柔性生产线工艺设计与准备过程管理,简化工艺准备流程,推进快换工装的应用并进行自动化程序管理;在柔性生产线加工运行过程中采用可视化管理方法,推动自动化环境下加工过程资源保障集中化管理。最终使该管理方法在某项目中进行成功实践,取得了良好的经济效益。     

基于深度学习的传感器优化布置方法

针对传感器优化布置(optimal sensor placement,简称OSP)问题,提出了一种新的使用深度神经网络的解决方案,并以简化的桥梁形状的桁架结构中的振动测试传感器优化为例进行了验证。首先,选择一种传统的传感器优化布置方法,对自动化生成的大量不同的桁架结构分别进行传感器优化布置计算,将所得优化布置结果在进行数据预处理后构建出深度学习方法所需要的训练集与验证集;其次,使用Python语言和深度学习框架TensorFlow设计实现与本研究问题适配的深度神经网络模型并训练;然后,随机生成了新的桁架结构参数;最后,将深度神经网络输出的传感器布置结果和传统方法的计算结果进行了比较,验证了本研究方法的有效性以及在速度上、可移植性与可扩展性方面的性能优势。     

Implementing lean practices in manufacturing SMEs: testing ‘critical success factors’ using Necessary Condition Analysis

Lean practices are known to increase operational performance. Previous research has identified critical success factors for implementing lean practices. This research aims to examine the extent to which success factors are critical for various degrees of lean practice implementation. Using multiple-respondent self-assessments from 33 Dutch manufacturing small and medium-sized enterprises (SMEs), we conducted a Necessary Condition Analysis. Our findings indicated that the criticality of success factors is progression dependent. In the initial stages of the lean journey, SMEs could improve their lean practices in a bottom-up manner through local factors such as a learning focus, improvement training and support congruence. When lean practices are more advanced, some company-wide factors must be present: top management support, a shared improvement vision and a supplier link. Our findings question the universality of success factors such as strategic involvement and indicate the need for a more dynamic model of lean implementation.     

Structural optimization with an automatic mode identification method for tracking the local vibration mode

Traditional optimization methods, which take a specific order of modal frequency as the design constraint, could fail to obtain the desired solution because of modal substitution. An improved optimization model with continuous sizing variables is established to solve this problem, in which the minimum weight and a given local modal frequency are considered as the objective and the constraint. To capture accurately the expected mode of vibration, a local mode identification technique is proposed based on the strain energy ratio between the local area and the whole structure. With that scheme, an optimization system is developed, in which the local mode can be effectively identified and the constraint can be updated with it in the iteration process. Two numerical examples, of a reinforced plate and a satellite structure, are applied to illustrate the effectiveness and efficacy of the proposed method.