Computation of an Emptiable Minimal Siphon in a Subclass of Petri Nets Using Mixed-Integer Programming

Deadlock resolution strategies based on siphon control are widely investigated.Their computational efficiency largely depends on siphon computation.Mixed-integer programming(MIP)can be utilized for the computation of an emptiable siphon in a Petri net(PN).Based on it,deadlock resolution strategies can be designed without requiring complete siphon enumeration that has exponential complexity.Due to this reason,various MIP methods are proposed for various subclasses of PNs.This work proposes an innovative MIP method to compute an emptiable minimal siphon(EMS)for a subclass of PNs named S⁴PR.In particular,many particular structural characteristics of EMS in S4 PR are formalized as constraints,which greatly reduces the solution space.Experimental results show that the proposed MIP method has higher computational efficiency.Furthermore,the proposed method allows one to determine the liveness of an ordinary S⁴PR.     

Efficient solution of the 3G network planning problem

The 3G universal mobile telecommunications system (UMTS) planning problem is combinatorially explosive and difficult to solve optimally, though solution methods exist for its three main subproblems (cell, access network, and core network planning). We previously formulated the entire problem as a single integrated mixed-integer linear program (MIP) and showed that only small instances of this global planning problem can be solved to optimality by a commercial MIP solver within a reasonable amount of time ( St-Hilaire, Chamberland, & Pierre, 2006). Heuristic methods are needed for larger instances. This paper provides the first complete formulation for the heuristic sequential method ( St-Hilaire, Chamberland, & Pierre, 2005) that re-partitions the global formulation into the three conventional subproblems and solves them in sequence using a MIP solver. This greatly improves the solution time, but at the expense of solution quality. We also develop a new hybrid heuristic that uses the results of the sequential method to generate constraints that provide tighter bounds for the global planning problem with the goal of reaching the provable optimum solution much more quickly. We empirically evaluate the speed and solution accuracy of four solution methods: (i) the direct MIP solution of the global planning problem, (ii) a local search heuristic applied to the global planning problem, (iii) the sequential method and (iv) the new hybrid method. The results show that the sequential method provides very good results in a fraction of the time needed for the direct MIP solution of the global problem, and that optimum results can be provided by the hybrid heuristic in greatly reduced time.     

基于混合整数规划的一般Petri网死锁检测方法

信标的受控性是检测柔性制造系统(flexible manufacturing system,FMS)Petri网模型是否存在死锁的关键因素.对于普通Petri网,在任何可达标识下所有信标不被清空是检测网系统非死锁的充分条件.然而,该条件对于建模能力更强的一般Petri网并不适用,max可控性条件由此产生.研究证明,该条件对于一般Petri网的死锁检测过于严格了.虽然其后有很多研究者通过改进max可控性条件以求给出条件更宽松的一般Petri网非死锁的充分条件,但大部分的研究成果都仅仅局限于一种顺序资源共享分配系统Petri网模型S4PR(systems of sequential systems with shared resources)网.因此,本文在max可控性条件的基础上提出了新的名为max#可控的信标可控性条件,并在此条件的基础上实现了基于混合整数规划(mixed integer programming,MIP)的死锁检测方法.与现有研究成果相比,max#可控性条件更宽松,可适用于更多类型的一般网,为解决大规模柔性制造系统中死锁监督控制器的结构复杂性问题提供了有力的理论支撑.     

Electrochemical copper (II) sensor based on self-assembled 4-amino-6-hydroxy-2-mercaptopyrimidine monohydrate

A 4-amino-6-hydroxy-2-mercaptopyrimidine monohydrate (AHMP)-based self-assembled monolayer (SAM) was formed on the gold electrode surface. Ellipsometric measurements evidenced the SAM formation on the gold electrode surface. The structural integrity of the modified gold electrode was also characterized by insulating properties of the SAM that were detected by cyclic voltammetry. The results of cyclic voltammetry showed that the SAM, which was formed by assembly of AHMP, was stable but did not completely block the redox-activity of ferrocene and K₃[Fe(CN)₆]/K₄[Fe(CN)₆]. In contrast completely blocked redox-activity was observed after the treatment of AHMP-based SAM with saturated solution of 4-formylphenylboronic acid in 1,4-dioxan. The modified electrodes exhibited a selective response towards Cu(II) ions in the presence of some interfering ions such as Cd(II), Co(II), Fe(II), Ni(II) and Pb(II). This study is the first scientific report on the application of AHMP-modified electrode as a selective Cu(II) sensor in the presence of some interfering cations.     

基于医学图像标准的最大深度投影算法

结合医学图像标准(DICOM)从最大深度投影(MIP)算法的基本原理、局限性和可能的改进方法等对现有的MIP技术进行评述，并从实际应用出发对MIP算法进一步的研究方向提出看法。     

A sensitive and selective sensor for dopamine determination based on a molecularly imprinted electropolymer of o-aminophenol

A simple and reliable method was proposed for preparing a selective dopamine (DA) sensor based on a molecularly imprinted electropolymer of o-aminophenol. The sensor is selective for the determination of DA in the presence of high concentrations of ascorbic acid (AA), with a maximum molar ratio of 1/1000. The molecular imprinted (MIP) sensor was tested by cyclic voltammetry (CV) as well as differential pulse voltammetry (DPV) to verify the changes in oxidative currents of ferricyanide. In optimized conditions, DA at concentrations of 2 × 10⁻⁸ to 0.25 × 10⁻⁶ mol/L could be determined with a detection limit of 1.98 × 10⁻⁹ mol/L (S/N = 3). The MIP sensor showed high selectivity, sensitivity, and reproducibility. Determination of DA in simulated samples of dopamine hydrochloride showed good recovery.     

一种MIP协议与micro-MIP协议相互融合的新方案

MIP协议为移动主机提供简单、扩展性强的全球移动，而各种micro-MIP协议则是解决域内快速、无缝切换的方案。当移动主机进行域间切换时，由于缺乏MIP协议与micro-MIP协议的协调机制将会导致移动主机的性能下降。提出了一种基于分组缓存和路由优化的宏观移动管理和域内micro-MIP协议无缝结合的新方案，可以提高All-IP网络的性能并保证移动主机的QoS性能。     

线性规划软件包GLPK的分析与应用

GLPK是一个求解大规模的线性规划问题(LP)、混合整数规划问题(MIP)以及相关问题的自由软件包。该文分析TGLPK的算法结构与数值计算等多方面的实现技术，并应用于解决NP-hard的调度问题。数值结果表明GLPK是研究LP和MIP问题强有力的工具。     

Molecularly imprinted ligand-exchange recognition assay of DNA by SPR system using guanosine and guanine recognition sites of DNA

Molecular imprinting polymer (MIP) as a recognition element for sensors are increasingly of interest and MIP-SPR have started to appear in the literature. In this study, we have proposed a novel surface imprinting method for DNA recognition and determination by methacryl-amidohistidine-platinum(II) [MAH-Pt(II)] as a new metal-chelating monomer via metal coordination–chelation interactions and guanosine/guanine templates of DNA. The study includes the measurement of binding interaction of molecular imprinted surface plasmon resonance (SPR) sensor via ligand interaction and recognition selectivity studies of guanosine and guanine imprinted polymers with respect to guanine and dsDNA. Adsorption isotherms of adenosine, guanosine, guanine, adenine, ssDNA and dsDNA were plotted and affinity constants for guanosine and guanine imprinted films were also calculated.     

基于位置预查询的因特网移动支持方案

提出一种预查询移动支持方案(mobile Internet protocol_based on location pre-query,简称MIP_Q),以解决家乡网络的流量瓶颈和单点故障问题,从而提高移动通信的效率和可靠性.MIP_Q通过扩展域名服务系统管理和跟踪移动节点的当前位置信息,省去了家乡代理;采用并行切换控制方式,同时避免了MIP(mobile Internet protocol,简称MIP)中的三角路由和隧道路由问题;借助有效的计算方法,分析和比较了MIP_Q与MIP,MIP_LR的平均移动通信成本和切换时延;在实现方面与广泛应用的广域蜂窝移动网络进行了类比.结果表明:MIP_Q在切换效率和新增实体数等方面优于同类方案;MIP_Q可以极大地降低节点的移动通信成本,减小切换时延;MIP_Q具有良好的可行性.最后提出MIP_Q的仿真和实现方案.     

分子印迹技术在水质检测传感器领域的应用

分子印迹技术是制备具有特异识别能力聚合物(分子印迹聚合物)的新兴化学合成技术,分子印迹聚合物可作为传感器识别元件,用于制备水质检测传感器。总结了分子印迹技术在水质检测传感器领域的应用研究现状,探讨了其发展趋势,指出以分子印迹聚合物作为敏感识别元件的水质检测传感器具有快速、准确、稳定和可反复利用等优点,可用于检测水体中农药、除草剂、神经毒剂以及无机离子化合物等,是实现水质现场快速检测的重要技术途径。     

结合环状原型空间优化的开放集目标检测

目的 现有目标检测任务常在封闭集设定中进行。然而在现实问题中,待检测图片中往往包含未知类别目标。因此,在保证模型对已知类检测性能的基础上,为了提升模型在现实检测任务中对新增类别的目标检测能力,本文对开放集目标检测任务进行研究。方法 区别于现有的开放集目标检测框架在检测任务中将背景类与未知类视为一个类别进行优化,本文框架在进行开放集类别识别的过程中,优先识别候选框属于背景类或是含待识别目标类别,而后再对含待识别目标类别进行已知类与未知类的判别。本文提出基于环状原型空间优化的检测器,该检测器可以通过优化待检测框的特征在高维空间中的稀疏程度对已知类、未知类与背景类进行环状序列判别,从而提升模型对开放集类别的检测性能。在（region proposal networks,RPN）层后设计了随机覆盖候选框的方式筛选相关的背景类训练框,避免了以往开放集检测工作中繁杂的背景类采样步骤。结果 本文方法在保证模型对封闭集设定下检测性能的情况下,通过逐步增加未知类别的数量,在Visual Object Classes-Common Objects in Context-20(VOC-COCO-20),Vi...     

移动IPRO中的MOBEE设计与实现

针对移动IP通信中的非优化路由问题,提出将主动网络技术植入几个关键功能实体,通过在MAC层设计一种移动IP执行环境MOBEE,对经过封装的主动信包做计算,使移动IP数据包主动化,以达到路由优化的目的。实验证明,大规模接入汇聚路由器ACR中基于硬件的MOBEE较Java.C等语言实现的执行环境有很大的性能提升。     

连铸-轧制生产中带有批决策的排序问题的建模与优化方法

以长材产线为背景, 研究了炉次在连铸及轧制阶段的组批及批排序问题. 与以往将连铸、轧制分开研究不同, 本文同时考虑连铸和轧制阶段对组批及批排序的要求, 还考虑了下游工序精整机组负荷均衡生产的要求. 为该问题建立了新的混合整数规划(Mixed integer programming, MIP) 模型. 由于问题的NP-hard 属性和模型的大规模特征, 以及工业应用的实际要求, 本文提出了改进的分散搜索(Scatter search, SS) 算法用于求解该问题. 在改进的SS 算法中, 利用解的相关性质来限制搜索空间, 并将变邻域搜索策略引入, 从而结合解的多样性及邻域互补性特点, 充分发挥算法混合的优势. 实际数据的计算结果验证了改进SS 算法的有效性.     

认知无线电中基于特征值检测的频谱感知算法

频谱感知的任务在于利用感知节点(无线传感器或者认知用户)采集的数据判断频谱空洞是否存在。基于最大特征值检测(MED)和最小特征值检测(SED)的方法最近被应用到频谱感知当中。这两种算法在检测实际应用当中普遍存在的相关信号时表现出良好的检测性能。然而,MED和SED算法对应的判决门限求解非常复杂,从而限制了它们在实际的认知无线电频谱感知中的应用。该文利用取样协方差矩阵的所有特征值,提出了一种新的基于特征值检测(ESD)的算法。利用多元统计理论获得了相应的判决门限。ESD算法无需主信号和无线信道信息参与感知过程。与此同时,它保留了与MED和SED相同的计算复杂度。更重要的是ESD算法对应的判决门限可以通过一个简单的闭合表达式进行求解,其计算复杂度低。仿真结果验证了新算法的有效性。     

Can mobile TV be a new revolution in the television industry?

The development within the mobile communications industry has provided a new mechanism for individuals to watch television programs via mobile device, known as mobile television (m-TV). Considering that m-TV is an emerging technology among traditional television services where critical success factors depend on user-centered, it is important to understand users' behavioral intention (BI). Unlike previous studies which have mainly focused on flow experience, perceived value, motivation and social cognitive theories, this study examines user-centered factors of technology readiness (TR) and technical support and training (TST). Therefore, 244 valid data were tested using partial least squares structural equation modeling (PLS-SEM) to measure both reflective and formative constructs based on the proposed framework. Findings reported that performance expectancy (PE), effort expectancy (EE) and TST have a positive and significant impact on BI. TR however has a negative relationship with BI. In addition, TR was also found to influence PE and EE. Similarly, TST has a direct effect factor on EE and EE was a significant determinant of PE. The findings of this study not only provide guidance for future researchers, but also valuable insights for designers, marketers and system providers of m-TV to tailor their services in the near future.     

Does value matters in playing online game? An empirical study among massively multiplayer online role-playing games (MMORPGs)

A few study examined the impact of value and aspects of behavioral intention in virtual environment. The aim of this study is to examine the impact of emotional value (VE), social value (VS), price-value for money (VP), performance-quality value (VQ) and repurchase intention (RI), willingness to pay a premium price (WTP) and word of mouth (WOM) among massively multiplayer online role-playing games (MMORPGs). A total of 228 valid questionnaires were collected from cybercafé customers in Klang Valley-Malaysia. Structural equation modeling (SEM) was employed using partial least squares (PLS) analysis to assess measurement and structural model for reflective construct. Our result reveals that there is a positive relationship between VP and RI, VQ and RI while there is no positive relationship between VE and RI, VS and RI. VE, VP and VQ value positively impact WOM but VS does not. VE and VP have positive while VQ and VS did not explain WTP. This study contributes to literature on the new phenomena of online game and is considered as few studies in examining value in Second Life setting. The practical and social implications of study are discussed along with research limitation and implication.     

A Novel Automatic Classification System Based on Hybrid Unsupervised and Supervised Machine Learning for Electrospun Nanofibers

The manufacturing of nanomaterials by the electrospinning process requires accurate and meticulous inspection of related scanning electron microscope (SEM) images of the electrospun nanofiber, to ensure that no structural defects are produced. The presence of anomalies prevents practical application of the electrospun nanofibrous material in nanotechnology. Hence, the automatic monitoring and quality control of nanomaterials is a relevant challenge in the context of Industry 4.0. In this paper, a novel automatic classification system for homogenous (anomaly-free) and non-homogenous (with defects) nanofibers is proposed. The inspection procedure aims at avoiding direct processing of the redundant full SEM image. Specifically, the image to be analyzed is first partitioned into sub-images (nanopatches) that are then used as input to a hybrid unsupervised and supervised machine learning system. In the first step, an autoencoder (AE) is trained with unsupervised learning to generate a code representing the input image with a vector of relevant features. Next, a multilayer perceptron (MLP), trained with supervised learning, uses the extracted features to classify non-homogenous nanofiber (NH-NF) and homogenous nanofiber (H-NF) patches. The resulting novel AE-MLP system is shown to outperform other standard machine learning models and other recent state-of-the-art techniques, reporting accuracy rate up to 92.5%. In addition, the proposed approach leads to model complexity reduction with respect to other deep learning strategies such as convolutional neural networks (CNN). The encouraging performance achieved in this benchmark study can stimulate the application of the proposed scheme in other challenging industrial manufacturing tasks.      

增强蒲公英算法优化乳腺癌图像多阈值分割

针对显微镜下乳腺癌病理组织图像结构复杂,细胞边界模糊等情况,基于传统的阈值分割在乳腺癌图像的分割应用中不能很好地实现把病灶区准确分离开来的问题,提出一种基于增强蒲公英优化算法(IDO)的乳腺癌图像多阈值分割方法.该方法引入IDO计算类间方差的最大值(Otsu)作为目标函数寻找最佳阈值, IDO建立回守策略解决传统蒲公英算法(DO)无限制搜索,超出像素范围的问题;引入对立式学习(OBL)避免算法陷入局部最优.实验结果表明,与哈里斯鹰算法(HHO)、人工猩猩部队优化算法(GTO)、传统蒲公英优化算法(DO)、海洋捕食者算法(MPA)相比,在相同阈值个数情况下IDO算法适应度值最大、收敛最快,并且在峰值信噪比(PSNR)、结构相似度(FSIM)、特征相似度(SSIM)这3个性能指标上也比其他对比算法更具有优势.     

融合CNN-BiLSTM-Attention的集成学习价格预测

价格预测对于大宗农产品市场的稳定具有重要意义,但是大宗农产品价格与多种因素有着复杂的相关关系.针对当前价格预测中对数据完整性依赖性强与单一模型难以全面利用多种数据特征等问题,提出了一种将基于注意力机制的卷积双向长短期记忆神经网络(CNN-BiLSTM-Attention)、支持向量机回归(SVR)与LightGBM组合的增强式集成学习方法,并分别在包含历史交易、天气、汇率、油价等多种特征数据的数据集上进行了实验.实验以小麦和棉花价格预测为目标任务,使用互信息法进行特征选择,选择误差较低的CNN-BiLSTM-Attention模型作为基模型,与机器学习模型通过线性回归进行增强式集成学习.实验结果表明该集成学习方法在小麦及棉花数据集上预测结果的均方根误差(RMSE)值分别为12.812, 74.365,较之3个基模型分别降低11.00%, 0.94%、4.44%,1.99%与13.03%, 4.39%,能够有效降低价格预测的误差.