CLUSTER OF WORKSTATIONS BASED ON DYNAMIC LOAD BALANCING FOR PARALLEL TREE COMPUTATION DEPTH-FIRST-SEARCH

IntroductionWith the advantages of the high- speed net-works,cluster of workstations achieve the samescalable parallelism as the current MPPs( Massive-ly Parallel Processors) .But there are many factorssuch as load imbalance,synchronization and com-munication that inhibit good performance in thesesystems.These issues are central to the develop-ment of many parallel applications. Load balanceproblem is a serious impediment to achieving goodperformance in parallel processing the issue of load…     

基于混合进化策略算法的并行多机调度问题研究

描述了一类带工艺约束的并行多机调度问题，其工艺约束的特点在于工件的加工需要多种不同类型机器同时协同工作来完成，为求解此类调度问题，采用进化策略与局部搜索相结合的混合算法，提出一种工件加工的机器动态调度策略，寻求生产调度完工时间和机器负荷均衡的优化。     

An Improved Model for Computing-Intensive Tasks on Heterogeneous Workstations

IntroductionIt is well known that many concurrent- com-puting tasks,which consume time badly,can besolved on a massive parallel system using parallelcomputing in order to acquire less execution time.Generally a massive parallel system is very expen-sive a…     

Parallel Distributed Implementation of Truss Optimization on a Local Area Network

A two-level optimization method for the design of complex truss and parallel distributed implementation on a LAN is presented using parallel virtual machine (PVM) for Win 32 as message passing between PCs. The volumes of truss are minimized by decomposing the original optimization problem into a number of bar optimization problems executed concurrently and a coordinate optimization problem, subject to constraints on nodal displacements, and stresses, buckling and crippling of bars, etc. The system sensitivity analysis that derives the partial derivatives of displacements and stresses with respect to areas are also performed in parallel so as to shorten the analysis time. The convergence and the speedup performances as well as parallel computing efficiency of the method are investigated by the optimization examples of a 52-bar planar truss and a 3 126-bar three-diraensional truss. The results show that the ideal speedup is obtained in the cases of 2 PCs for the 3 126-bar space truss optimization, while no speedup is observed for the 52-bar truss. It is concluded that (1) the parallel distributed algorithm proposed is efficient on the PC-based LAN for the coarsegrained large optimization problem; (2) to get a high speedup, the problem granularity should match with the network granularity;and (3) the larger the problem size is, the higher the parallel efficiency is.     

基于EBE的Bi-CG迭代法求解Navier-Stokes方程的并行计算

阐述了用有限元求解流场不可压Navier-Stokes方程时，基于单元接单元（Element-By-Element,EBE）技术的并行双共轭梯度法(Bi-CG)。在国家高性能计算中心（成都）的曙光2000并行系统上，采用MPI消息传递实现了此种算法，计算定常三维空穴水域流场，取了较好的加速比。由于不需存储系数矩阵，在很大程度上节省了内存，大大提高了求解流场问题的规模。     

前言

基于RP(Revealed Preference)和SP(Stated Preference)调查数据，利用潜在类别模型对高铁旅客进行细分，得到旅客对平行车次不同服务属性，如列车运行时间、发车时段和舒适度的偏好程度，并对其进行量化；引入收益管理，以多列车整体收益最大为目标，构建平行车次动态差别定价模型，并设计模拟退火算法进行求解；最后，通过京沪高铁进行实例验证.结果表明：与固定票价进行客票销售相比，所提方案能够适应高峰期和平峰期不同客流特点，提高铁路客票总收益，为高铁平行车次灵活定价提供参考.     

考虑收益管理的高铁平行车次动态差别定价

基于RP(Revealed Preference)和SP(Stated Preference)调查数据，利用潜在类别模型对高铁旅客进行细分，得到旅客对平行车次不同服务属性，如列车运行时间、发车时段和舒适度的偏好程度，并对其进行量化；引入收益管理，以多列车整体收益最大为目标，构建平行车次动态差别定价模型，并设计模拟退火算法进行求解；最后，通过京沪高铁进行实例验证.结果表明：与固定票价进行客票销售相比，所提方案能够适应高峰期和平峰期不同客流特点，提高铁路客票总收益，为高铁平行车次灵活定价提供参考.     

Outlier mining based abnormal machine detection in intelligent maintenance

Assessing machine’s performance through comparing the same or similar machines is important to implement intelligent maintenance for swarm machine. In this paper, an outlier mining based abnormal machine detection algorithm is proposed for this purpose. Firstly, the outlier mining based on clustering is introduced and the definition of cluster-based global outlier factor (CBGOF) is presented. Then the modified swarm intelligence clustering(MSIC) algorithm is suggested and the outlier mining algorithm based on MSIC is proposed. The algorithm can not only cluster machines according to their performance but also detect possible abnormal machines. Finally, a comparison of mobile soccer robots’ performance proves the algorithm is feasible and effective.     

一种少自由度并联机器人雅克比分析的新方法

提出一种对称少自由度并联机器人雅克比方阵的建立方法.利用螺旋理论分析单个支链的结构,找出动平台对驱动关节的映射系数,根据不同的输入选取,对映射系数进行组合,去除映射系数矩阵中非独立运动参数对应的列,得到满秩的映射系数矩阵.对映射系数矩阵求逆,得到对称少自由度并联机器人的雅克比方阵,说明了运动输入与运动输出之间的对应关系,并得到实例验证.     

并行机器人结构刚度有限元数值模拟

通过有限元数值模拟对并行机器人的结构刚度进行了分析。并行机器人的刚度可以通过有限元数值模拟方法对其进行分析．基于子结构的概念并鉴于大多数并行机器人都具有结构对称性特征。提出了一种有效的数值模拟方法——模块设计方法，并通过一个应用实例显示了这种方法的有效性．对数值模拟结果进行了整理并在此基础上绘制了Isoglide3-T3并行机器人的刚度图。     

两自由度并联机器人的RBF神经网络辨识自适应控制

针对平面两自由度五杆并联机器人的轨迹跟踪问题,提出了一种基于RBF神经网络的自适应PID控制方法.该控制方案利用RBF神经网络自适应学习辨识并联机器人系统的未知非线性动态,可以在线调整PID控制参数以实现高精度控制.仿真结果显示该控制策略可以精确实现对于并联五杆机器人的轨迹跟踪控制,该方法的自适应性和跟踪性能均优于传统的PID控制.     

产业集群并行供应链间的企业合作决策分析

引用博弈理论的方法,讨论在产业集群并行供应链中处于相似价值链位置的核心企业之间的竞争与合作关系,得出合作优于竞争的结论。同时,在此基础上提出产业集群中第三方组织商会的概念,商会通过监督和采取一系列激励机制保障集群企业在一个稳定的环境中持续、健康的发展。     

平面二自由度冗余驱动并联机器人控制实验研究

对平面二自由度冗余驱动并联机器人进行了逆运动学分析,由末端执行器的位置计算得到主动关节角的位置.分别采用传统的PID控制方法和PID+速度前馈补偿+加速度前馈补偿的方法对并联机器人轨迹跟踪问题进行了实验研究.通过比较2种控制方法的轨迹跟踪效果和比较由连接在连杆上的加速度传感器测量的振动大小,证明了PID+速度前馈补偿+加速度前馈补偿的方法能更好地实现轨迹跟踪.     

Normalized Design of Multi-channel LIF Digital Receiver

IntroductionThe consistency of amplitude and phase be-tween the two orthogonal outputs will be improvedconsiderably by adopting the LIF digital receiverwith parallel processing design[1,2 ] .Meanwhile,LIF digital processing leads to a low amount of re-al- time data processing.As a result,a multi- chan-nel LIF digital receiver can be implemented usingtime division multiplex and a parallel processingarray( constructed by high speed DSP chips) ,suchas Sharc 2 1 0 6x and TMS32 0 C6x,and thu…     

Facility Layout Based on Sequence Analysis： Design of Flowshops

A computer-aided method to design a hybrid layout—tree-shape planar flowlines is presented. In newtype flowshop layout, the common machines shared by several flowlines could be located together in functional sections. The approach combines traditional cell formation techniques with sequence alignment algorithms. Firstly, a sequence analysis based cell formation procedure is adopted; then the operation sequences for parts are aligned to maximize machines adjacency in hyperedge representations; finally a tree-shape planar flowline will be obtained for each part family. With the help of a sample of operation sequences obtained from industry, this algorithm is illustrated.     

Grid Service Framework： SupportingMulti-Models Parallel Grid Programming

IntroductionVolunteer computing has the idea that homePCs are mostly idle and thus could be harnessed forsolving complex computational problems.Thisidea is to decompose the problem into many chunksthat can run concurrently with very little interac-tion,referred by some as“embarrassingly parallel”computation.The best- known example is SETI@HOME.A comparison of traditional high per-formance server versus volunteer computing isshown in Fig.1 .Though volunteer computingmodel has been succes…     

基于旅客选择行为的高速铁路平行车次定价策略

考虑运行在同一OD上的高速铁路平行车次,现行高铁票价固定,即不能反应平行车次间的差异化,也无法与旅客需求状况相适应,最终造成平行车次间上座率的失衡和收入的损失.鉴于此,本文借鉴多产品定价的思想,利用MNL模型来描述旅客购票过程的选择行为,并以平行车次的整体收入最大化为目标提出了3种定价策略.最后以京沪高铁为案例背景,对旅客购票过程的最优票价路径及3种定价策略的效率进行了分析.结果表明,3种定价策略均能有效提高售票收入,且差别动态定价策略表现最优.本文研究内容为高速铁路平行车次的定价决策提供了科学依据.     

Application of PETSc in soil-water coupled geotechnical problems

In this paper, parallel library, portable extensible toolkit for scientific computation (PETSc), is used to solve linear systems in soil-water coupled finite element method (FEM) for geotechnical problems. The parallel environment is integrated into GLEAVES, which is a geotechnical software package used for the finite element simulation. The linear system ·· = · which is a fundamental and the most time-consuming part of the FEM is solved with iterative solvers in PETSc. In order to find a robust and effective combination of iterative solvers and corresponding preconditioners for the soil-water coupled problems, performance evaluations on Krylov subspace methods and four preconditioners are carried out. The results indicate that general minimal residual (GMRES) method coupled with preconditioners can provide an effective solution. The application to a construction project is presented to illustrate the potential of the proposed solution.     

A New Parallel Algorithm in Power Flow Calculation: Dynamic Asynchronous Parallel Algorithm

Based on the general methods in power flow calculation of power system and onconceptions and classifications of parallel algorithm, a new approach named DynamicAsynchronous Parallel Algorithm that applies to the online analysis and real-time dispatching and controlling of large-scale power network was put forward in this paper. Its performances of high speed and dynamic following have been verified on IEEE-14 bus system.     

���ٹ�·��ƽ�и�·��ͨ���ֵ�����Ӱ�����ط���

在公路可行性研究或项目评估时，同一通道内高速公路与平行辅路交通量分担比例在交通量预测中非常重要，本文以交通运输的基本机理和追求的目标为分析的出发点，系统的分析了交通运输行为主体和公路属性的各方面对该比例的影响，包括地区的经济水平、不同车型的差异性、高速公路与平行辅路的里程差、等级和路况的对比、通道的交通饱和程度、收费水平等因素，这些因素的分析对实践中该比例的确定具有较强的指导意义。