支持广域网的协作式服务发现机制

服务发现是网络分布式环境下进行信息共享、数据集成、流程协作的前提。广域网下的服务发现机制必须在无可直接利用的广播和组播机制上解决系统的规模伸缩性问题,以合理的代价为用户提供高效的服务发现机制。采用覆盖网络体系结构,并实际构建了具有一定可伸缩性的服务发现系统Service CatalogNet。Service CatalogNet基于分布存储的服务信息和后缀树形式的服务信息索引,实现了协作式服务发现机制,特别提供了QoS感知的服务发现机制,即可基于客户的QoS指标生成应用层选播/多播路由,进而实现服务发现。实验结果表明,该服务发现机制在发现性能上优于LDAP实现。

A double wavelength DFB laser with an identical active area for CWDM

A novel distribute feedback (DFB) laser which gave two different wavelengths under two distinct work conditions was fabricated. The laser consists of two Bragg gratings with different periods corresponding to wavelength spacing of 20 nm in an identical active area. When driving current was injected into one of the different sections separately, two different wavelengths at 1542.4 and 1562.5 nm were realized. The side mode suppression ratio (SMSR) of 45 dB or more both for the two Bragg wavelengths were achieved. The fabricating process of the laser was just the same as that of traditional DFB laser diode. This device can be potentially used in coarse wavelength division multiplexer (CWDM) as a promising light source and the technology idea can be used to enlarge the transmission capacity in metro area network (MAN).

集群网络评测模型的新探索

传统集群网络(cluster area network,简称cLAN)的评测模型主要考虑了延迟、带宽、路由、拥塞、网络拓扑结构等因素.但这些因素是否足以描述实际应用程序在集群上的通信行为,或者对其在集群系统上的性能给出一个很好的预测呢?当对NAS Parallel Benchmark(2.4版本)在集群系统深腾1800(DeepComp 1800)上进行大量测试时发现,集群网络的通信性能可以被一种特殊的通信模式(LU模式)所严重影响.更深入的研究表明,这个影响LU模式的因素是独立于前面所述的如延迟、带宽、路由、拥塞、网络拓扑结构等因素的.因此有必要对集群网络的评测模型重新进行审视,并增加一个新的性能评测因子以反映这个新发现的现象.从研究结果来看,这个重新审视也将对集群系统上的并行算法设计以及实际大规模科学计算的应用程序性能的优化提供一些新的思路.

适用于云计算的面向查询数据库数据分布策略

为满足海量数据的处理需求,业界提出了多种解决方案。云计算是目前较为热门的一种,它主要用廉价PC组成超大规模集群服务器来进行数据存储和处理。随着云计算技术的发展,越来越多的应用将转移到云中,数据库系统也不例外。但数据库系统要求的ACID特性在数据分布存储时可能导致部分操作性能低下,如连接查询操作。为在数据分布存储下提高数据库系统的性能,提出了一种面向查询的数据分布策略(Selection Oriented Distribution,SOD),即根据数据库的查询情况确定数据的分布算法。该算法适用于云计算,能明显提高系统的查询性能。

RH6-B机器人的通信协议

RH6-B机器人控制器是基于CAN总线的控制器,具有CAN总线的接口。实现了多机器人监控系统,并开发了相应的协议和通信系统。该文主要介绍了它的系统结构、总线协议和相应的软件,并对性能进行了分析。

星球探测系统的多机器人组内协作研究

建立了子机器人的控制模型,提出了分散式控制与集中式控制相结合的复合控制体系。采用动态分配ID号和由ID号确定机器人组中领导者的机制,建立了有领导者的机器人组的协作方法。其中,领导者与监控平台之间采用无线通讯,机器人组内采用CAN(control area network)总线传递控制信号。同时采用组内基于状态表匹配的控制方式。两个机器人组合利用以上机制,通过相互协调完成差速转弯的实验验证了该控制方法的可行性。

仿人机器人控制系统的研究与实现

根据仿人机器人控制性能的要求,设计开发了关节控制器,并通过CAN总线把各个关节控制器、力传感器及上位机连接在一起,构成了分布式控制系统.利用无线局域网技术,实现了语音、视频等多媒体信息的传输,把监控台、头部、上身和移动平台连接在一起,构成了仿人机器人完整的控制系统.最后提出了一些设想以提高系统的性能.

AUV内部通讯总线设计

设计了基于CAN协议的AUV内部通讯总线系统 .系统通过协议转换器的模块化、可配置性设计满足AUV系统对其内部通讯总线的开放性要求 .协议转换器内部的容错处理能力以及紧急事件处理节点的设计为增强AUV系统的可靠性和容错能力、为避免AUV在深海工作环境下丢失增加有力的保障措施

基于CAN总线的分布式数据采集与控制系统

随着电子技术和计算机技术的不断发展，工业生产过程的控制系统正在向着智能化、数字化和网络化的方向发展。传统的集散控制方式和计算机分层控制方式已经开始让位于智能终端与网络结合的总线网络控制方式。当今，在工厂中过程控制环境下的分布式自动化系统变得越来越复杂，尤其系统内部的各设备之间需要快速交换大量的信息，以便实现对被控系统更为精确的控制和提供一些辅助的评价函数。这就意味着要不断增加带宽和提高通信速率以满足网络通信的需要。在现有的多种可利用网络设备中，CAN总线以其清晰的定义、极高的可靠性及其独特的设计，被认为是最能有效地解决这一问题的途径之一。而且市场上基于通信技术的产品中，就实时性考虑，由于CAN总线采用的非表意性的通信方式，因此其结构更为简单，实时性更好。基于此背景，我们以CAN总线作为通信媒介，将分布于各控制现场的传感器、执行器和控制器有序地连接起来，构成了一个基于CAN总线的分布式局域网络控制系统。本文首先介绍了基于CAN总线的分布式数据采集与控制系统的总体结构。然后从硬件方面描述了基于CAN总线的通信协议转换单元、数据采集单元和输出控制单元的功能、硬件配置及各单元功能的具体实现过程，给出了各单元的性能指标。软件方面，以C语言作为平台，开发了基于CAN总线的上位计算机管理与监控软件，实现了对整个网络设备的系统管理和系统控制功能。对于该总线系统，作者运用了PID控制和模糊控制算法实现了对水箱液位的控制，达到了理想的效果。基于CAN总线的控制系统很好地解决了集散控制系统难以解决的难题，模糊控制的应用能很好地把总线控制系统应用到具有非线性、大时滞和难于获得精确模型的控制系统中。

濮城油田沙三中6-10高分辨率层序地层学研究

The foundation of reservoir model and residual oil prediction have been the core of reservoir detailed description for improved oil production and enhanced oil recovery. The traditional way of sandstone correlation based on the geometrical similarity of well-logs which emphasizes "based on the cycle and correlating from larger to smaller" has shown its theoretical limits when explaining the correlating and the scale, geometry, continuity, connectivity of sandstones and the law of the reservoir property. It has been an urgent and difficult subject to find new theory and methods to solve the reservoir correlation and property prediction. It's a new way to correlate strata and found framework of reservoir through the process-response analysis in the base-level cycles. And it is also possible to analyze the reservoir property in reservoir framework. Taking the reservoir of zonation 6-10 in S3~2 of Pucheng Oil Field in Henan Province as an example, we founded the detailed reservoir stratigraphic framework through base-level correlation. In the strata frame, sediment distribution and its development are discussed based on sediment volume partitioning and facies differentiation analysis. Reservoir heterogeneities and its relation to base-level are also discussed. The analysis of primary oil distribution shows the base-level controlled oil distribution in reservoir. In this paper, subjects as following are discussed in detail. Based on the analysis of sedimentary structure and sedimentary energy, the facies model was founded. Founding stratigraphy framework through base level analysis In the studying zone, one long term cycle, 6 middle term cycles and 27 short term cycles was identified and correlated. 3 Predicting the property of reservoir for improving oil development The base level controlled the property of sandbody. The short and very short term cycle controlled the pattern of heterogeneities in sandbody, and the middle and long term cycle controlled the area and inter-layer heterogeneities. On the lower location of the middle and long term base level, the sandbody is well developed, with a wide area and large thickness, while on the high location of base level, there is an opposite reservoir character. 4 The studying of reservoir development response and oil distribution making a solid base for development adjustment Primary oil distribution is controlled by base level location. It tells that the sandbody on the high base level location was poor developed for its difficulty to develop. While on the low location of the base level, the sandbody is well developed for its relative easy to develop and dominant role in the development, but high residual oil for its high original oil content.