含风险时差的工序可间断多资源均衡分析

针对网络计划资源均衡问题,已经提出过一种工序可间断的单资源均衡模型。通过引入多资源的条件和风险时差的概念,对该模型加以改进。改进后的模型在求解多资源均衡问题的同时,可以为某些选定的工序预留出一定的时差,从而避免因资源均衡过程对这些时差的过度利用而增加工期风险。通过采用遗传算法对模型进行求解,验证了模型的有效性。     

新型智能算法解决隧道工程中的NP-hard问题

针对工作连续性与资源恒定性两种约束对电力隧道工程的不利影响,笔者建立了一个工作可间断且资源可波动的离散时间费用权衡模型(DTCTP-wr),并通过设计一种双链式整数编码和随机单点交叉算子的改进遗传算法进行求解。此外,通过一个实际的电力隧道建设项目验证该算法的有效性,并设置四种情景对工作可间断与资源可波动进行分类分析。研究结果表明：工作间断能够得到比工作连续时总工期更短的方案;资源可波动能够得到比资源恒定时总费用更少的方案;同时允许工作间断与资源波动不仅能够让施工方得到更多可行的方案而且能够为管理者提供总工期和总费用双目标优化的调度方案。     

带有间断系数椭圆方程的加权间断Galerkin方法

本文应用间断Galerkin(DG)方法求解带有间断系数的二维椭圆方程.针对扩散系数间断的特点,我们构造一种新的加权对称内惩罚方法.证明了相应双线性形式的连续性和强制性,并给出收敛性证明.数值算例表明我们的DG方法对于求解强间断系数问题十分有效.     

资源获胜率保证的网格任务调度算法

为保证多资源竞争下单个资源的获胜率,建立一个资源富裕型的网格计算环境模型,基于此模型提出了一个基于博弈论的资源竞争获胜率保证算法（GVP）。GVP算法通过对已知信息分析,预测对手在下一次博弈中的行动,并在此基础上确定自己的行动,使资源累计的获胜率维持在一个可接受的范围。通过实验比较了使用不同竞争策略资源的获胜率。对实验结果进行了深入分析,讨论了两资源博弈的纳什均衡解。     

基于Hermite插值的SVM研究

在传统SVM的分类求解算法中,由于严格凸的无约束最优化问题中单变量函数x+是不可微的,不能使用通常的最优化的算法进行求解。三次Hermite插值多项式光滑的支持向量机模型采用的是一种多项式光滑技术,用三次Hermite插值多项式代替单变量函数x+,将原来不可微的模型变为可微的模型,并且给出了三次Hermite插值多项式光滑化单变量函数x+的推导过程。使用UCI机器学习数据集中的数据,通过实验验证了该模型的有效性。     

具有间断系数热传导方程的局部间断Galerkin方法

本文给出了求解具有间断系数热传导方程稳定的局部间断Galerkin方法.理论表明,当采用k阶多项式有限元空间逼近时,该方法连续时间的L~2模误差估计阶为O(h~(k+1/2)).文中分别应用显式和隐式时间离散求解局部间断Galerkin格式,数值算例验证了方法的有效性和理论结果.     

主动攻击下的隐写系统博弈模型

针对隐写系统面临的主动攻击问题,对隐写方和主动攻击方之间的对抗关系进行建模,提出了以信息嵌入率和错误率两个目标为收益函数的隐写系统博弈模型。借助二人有限零和博弈基本理论,分析了隐写方和主动攻击方博弈均衡的存在性,并给出了均衡局势下对抗双方的策略求解方法。最后通过求解一个实例说明了模型的有效性。建立的模型可为隐写方和主动攻击方的最优策略选择提供理论依据,对抗主动攻击的隐写算法设计也具有一定的指导意义。     

主动攻击下的隐写系统博弈模型

针对隐写系统面临的主动攻击问题,对隐写方和主动攻击方之间的对抗关系进行建模,提出了以信息嵌入率和错误率两个目标为收益函数的隐写系统博弈模型。借助二人有限零和博弈基本理论,分析了隐写方和主动攻击方博弈均衡的存在性,并给出了均衡局势下对抗双方的策略求解方法。最后通过求解一个实例说明了模型的有效性。建立的模型可为隐写方和主动攻击方的最优策略选择提供理论依据,对抗主动攻击的隐写算法设计也具有一定的指导意义。     

具有QoS保证的流媒体业务服务组合算法*

针对采用服务组合方式实现流媒体业务提供过程中的服务路径选择问题,从物理资源和逻辑资源两个角度出发,选择时延、带宽、丢包率和服务实例可用服务能力作为参数,建立了多约束归一化线性模型;同时,根据模型特点,采用线性整数规划迭代求解,提高了处理效率。仿真实验证明,该模型和算法能够有效地保障业务服务质量,并具备良好的负载均衡效果。     

两车间可调度工序均衡处理的综合调度算法

在两车间具备相同设备资源的生产条件时,需要考虑产品完成时间和车间之间工序移动次数尽可能少的问题。为此,提出两车间可调度工序均衡处理的综合调度算法。为减少单件复杂产品的完成时间,针对可调度工序的灵活性、并行性和两车间设备相同的条件,采用可调度工序车间均衡策略进行分组。为减少工序移动次数,按分组工序车间确定策略分配工序所在车间,并进行调度。实例结果表明,该算法可实现两车间综合调度,且产品完成时间和车间之间的工序移动次数较少。     

Symmetry Breaking Constraints for Value Symmetries in Constraint Satisfaction

Constraint satisfaction problems (CSPs) sometimes contain both variable symmetries and value symmetries, causing adverse effects on CSP solvers based on tree search. As a remedy, symmetry breaking constraints are commonly used. While variable symmetry breaking constraints can be expressed easily and propagated efficiently using lexicographic ordering, value symmetry breaking constraints are often difficult to formulate. In this paper, we propose two methods of using symmetry breaking constraints to tackle value symmetries. First, we show theoretically when value symmetries in one CSP correspond to variable symmetries in another CSP of the same problem. We also show when variable symmetry breaking constraints in the two CSPs, combined using channeling constraints, are consistent. Such results allow us to tackle value symmetries efficiently using additional CSP variables and channeling constraints. Second, we introduce value precedence, a notion which can be used to break a common class of value symmetries, namely symmetries of indistinguishable values. While value precedence can be expressed using inefficient if-then constraints in existing CSP solvers, we propose efficient propagation algorithms for implementing global value precedence constraints. We also characterize several theoretical properties of the value precedence constraints. Extensive experiments are conducted to verify the feasibility and efficiency of the two proposals.     

基于神经网络预测模型输入参数配置方法的实现

基于数据挖掘中的关联概念,提出了一种针对神经网络预测模型训练参数的选择方法,有效地提高了神经网络模型在毛纺工艺中对纱线断头率的预测精度;该方法通过生产中的训练参数记录进行关联规则的提取,可快速的排除产生负面影响的训练参数,迅速选择可以提高预测精度的训练参数,从而达到提高神经网络模型预测性能的目的;实验证明,利用关联算法进行参数配置,可以有效提高神经网络输入模型的预测精度.     

Breaking the rules in interactive media design education

In today's interactive media design it is difficult for a designer to create aesthetic innovations and to break free from ordinariness. The most important factor limiting interactive media design aesthetics is that education seems to be more focused on following traditional rules of interaction design rather than innovative approaches. These rules limit creativity and often relegate design students to producing ordinary interface solutions. This is especially burdensome for us as teachers. In order to address this problem, we developed an education model inspired by Lars von Trier's film Five Obstructions. We call this model ‘breaking the rules’. In the ‘breaking the rules’ approach students produce, within a range of probabilities, design problem solutions in cases of total or partial visual/auditory/tactile obstructions. The most important outputs of the model are (1) to make design student think/look outside of the ordinary, (2) to produce unusual solutions, (3) to maximise design solutions with sound.     

基于支持向量机的Free}ate软件流量检测研究

近年来出现了可以突破网络过滤访问国外被禁止信息的破网行为。针对破网行为的研究与控制,具有十分 重要的现实意义。流量分类技术一直是国内外网络测量方向的研究热点,并在P2P检测领域中取得了很好的效果。 将流量分类领域中的支持向量机技术应用于破网软件frccgatc的行为检测。实验结果表明,该方法对于破网行为产 生的流量具有较高的检测率,为有效监测破网行为提供了一种新思路。     

High‐quality pagination for publishing

The problem of line breaking consists of finding the best way to split paragraphs into lines. It has been cleverly addressed by the total‐fit algorithm exposed by Knuth and Plass in a well‐known paper. Similarly, page‐breaking algorithms break the content flow of a document into page units. Formatting languages—such as the World Wide Web Consortium standard Extensible Stylesheet Language Formatting Objects (XSL‐FO)—allow users to set which content should be kept in the same page and how many isolated lines are acceptable at the beginning/end of each page. The strategies most formatters adopt to meet these requirements, however, are not satisfactory for many publishing contexts as they very often generate unpleasant empty areas. In that case, typographers are required to manually craft the results in order to completely fill pages. This paper presents a page‐breaking algorithm that extends the original Knuth and Plass line‐breaking approach and produces high‐quality documents without unwanted empty areas. The basic idea consists of delaying the definitive choice of breaks in the line‐breaking process in order to provide a larger set of alternatives to the actual pagination step. The algorithm also allows users to decide the set of properties to be adjusted for pagination and their variation ranges. An application of the algorithm to XSL‐FO is also presented, with an extension of the language that allows users to drive the pagination process. The tool, named FOP+, is a customized version of the open‐source Apache Formatting Objects Processor formatter. Copyright © 2011 John Wiley & Sons, Ltd.     

基于OpenGL的飞机刹车视景仿真系统研究

视景仿真作为飞机刹车半实物仿真的一部分,实现了飞机刹车仿真的可视化.在设计过程中将OpenGL、3DS Max和串行通信技术有效结合,解决了飞机刹车过程的三维视景系统模拟仿真问题.仿真软件以Visual C 为平台,采用开放型图形库(OpenGL)实现了飞机刹车过程可视化的场景绘制、纹理映射、视点变换及动画仿真,利用3DS Max建立了飞机实体模型,飞机刹车过程的动态演示随RS-485实时通信数据不断刷新,以实现飞机刹车过程的可视化处理、数据记录和视景回放.     

Animation of fracture by physical modeling

The breaking of solid objects, like glass or pottery, poses a complex problem for computer animation. We present our methods of using physical simulation to drive the animation of breaking objects. Breakage is obtaned in a three-dimensional flexible model as the limit of elastic behavior. This article describes three principal features of the model: a breakage model, a collision-detection/response scheme, and a geometric modeling method. We use networks of point masses connected by springs to represent physical objects that can bend and break. We present effecient collision-detection algorithms, appropriate for simulating the collisions between the various pieces that interact in breakage. The capability of modeling real objects is provided by a technique of building up composite structures from simple lattice models. We applied these methods to animate the breaking of a teapot and other dishware activities in the animationTipsy Turvy shown at Siggraph '89. Animation techniques that rely on physical simulation to control the motion of objects are discussed, and further topics for research are presented.     

基于粒子滤波算法的高速公路车辆停车检测

提出了一种基于粒子滤波视频跟踪算法的停车事件检测方法,实现了对高速公路交通视频的自动监控。首先用混合差分技术,快速提取出视频中的车辆对象;并用粒子滤波算法实现了运动车辆的跟踪;进而通过对车辆运动的数学建模,对停车事件进行了自动检测。最后,对多组高速公路交通视频进行测试,结果表明：提出的检测方法比其他常用方法响应速度更快,且具有较高的检测准确率和鲁棒性。     

89C52在电缆断点测定系统中的应用

介绍了一种利用低压脉冲反射法测定电缆断点位置的方法,并采用89C52单片机建立了控制系统。通过发射低压脉冲波和接收反射的低压脉冲波,实现对电缆中断点位置的自动测定。该系统结构简单,应用效果良好。