NoC 中基于数据包分类的功率门控策略

随着制造工艺的精进,静态功耗逐渐成为路由器的主要功耗来源之一。 针对基于片上网络架构的多核系统高功耗问 题,提出了一种基于数据包分类的功率门控策略,并修改了路由器架构。 首先对数据包进行分类,其次利用旁路对分类数据包 进行不同的处理,来绕过休眠路由器,从而降低由于功率门控的应用而导致的额外数据包延迟和功耗。 旁路可以保证路由器休 眠时的数据包传输,增加休眠路由器的休眠时间,减少静态功耗。 数据包的分类处理使得路由器的休眠和唤醒更加合理高效。 仿真结果表明,本文方案能明显减少路由器的静态功耗和网络的延迟,相比传统的功率门控技术,最大减少 72. 4%静态功耗和 16. 8%平均数据包延迟;相比经典的采用功率门控的旁路路由器,减少 12. 4%静态功耗和 4. 7%平均数据包延迟。 路由器的额 外硬件开销仅增加了 3. 2%。     

WiNoC中协同拥塞控制的高性能路由器设计

无线路由器作为无线片上网络中无线通信的关键部件,若出现拥塞将会严重影响网络性能。基于此,提出一种协同拥塞控制的路由器(C~3WR)设计方案。该设计针对无线接口接收端缓冲区拥塞,首先通过无线输入端口与本地其他输入端口缓冲区级联的方式,为拥塞的数据提供多路径传输,其次提出WCC算法分流数据包缓解无线接收端拥塞。针对无线接口发送端缓冲区拥塞,采用动态路由策略,降低对拥塞无线路由器的流量注入率。实验结果表明,相较于对比对象,C~3WR路由器吞吐率最多增加33.80%,网络延时最多降低33.17%,面积功耗开销可接受。该方案以协同拥塞控制的方式均衡网络流量分布,保证了系统的性能又充分利用了可用资源。     

基于关键IP核加固的片上网络容错机制

片上网络中的路由器故障将导致与其相连的IP核不能通信,严重影响了片上网络的性能.因此提出一种基于片上网络2D-mesh结构的容错机制,通过将关键IP核的资源网络接口与相邻节点的资源网络接口相连进行IP核的加固,在每个路由器的各个端口中配置邻居节点状态寄存器标示邻居节点的好坏,在路由计算时通过检查寄存器绕过故障路由器,同...     

Torus拓扑结构的双端口NoC模型与性能分析

为了充分利用片上网络的并行通信能力,挖掘网络带宽利用率,设计了一种高性能的双端口片上网络。该网络中每个路由节点提供两个本地端口,每个资源节点采用对角线的方式接入到网络中两个路由节点上,并针对这种接入结构提出一种目的节点切换方法用于提高网络的并行通信能力。同时,为进一步提高新路由结构的性能,将双端口路由节点引入Torus结构,构造了基于Torus的双端口网络的模型,并对其进行性能评估。实验结果表明,基于Torus结构的对角线双端口NoC在单目事物实验中较单端口网络平均吞吐量和平均包延迟最大改善了83.3%和55.9%,在双目事物实验中较同一维序双端口网络平均吞吐量和平均包延迟最大改善了91.1%和54.3%。     

3D NoC中柔性可配置的高可靠路由器设计

三维片上网络中路由器发生故障及拥塞等可靠性问题,会影响整个网络性能。因此针对路由器输入缓存的故障和拥塞问题,提出一种柔性(flexible)可配置的高可靠路由器架构。每条输入链路和2个相邻的输入缓存相连。通过建模,根据具体的故障和拥塞情况,选择合适的输入缓存路径,实现部分缓存的共享。不仅能达到路由器故障的容错目的,还能在网络重负载的情况下有效的解决网络拥塞问题。实验结果表明,方案相较于传统路由器方案,在一般传输模式和0.5 filts/node/cycle的注入率下,无故障时平均时延下降了81.89%,2个数据分配器故障时平均时延下降了87.38%。在网络出现故障和拥塞时,方案具有明显的优势,很好的保证了整个网络的高可靠性以及低时延。     

基于时间复杂度的无线网络编码数据包传输次数优化分析

为了提高无线广播网络传输的稳定性,建立了一种基于时间复杂度的无线网络编码数据包传输次数优化网络模型。分析结果表明:随着节点数M不断增大,将获得更高的数据包传输次数。NCIF方案具有比CLIF更优的性能,并且比SRPSiF的性能也更好,采用NCIF方案可以获得更高的重传效率。NCIF方案具备比SRPS方案更优的性能,NCIF方案利用S生成编码包的过程中不考虑中继节点R产生的解码性,可以高效恢复目的节点的丢失数据包。当丢包率增大后,NCIF与CLIF方案都发生了性能降低的情况,降低了中继节点的作用。     

基于故障粒度划分的NoC链路自适应容错方法

No C中数据包的传输需要经过链路,且链路占据了芯片上的绝大部分面积,故对链路的容错尤为重要.为解决链路的故障问题,该文将链路故障进行粒度划分,并提出了一种自适应的链路容错方法.当链路出现粗粒度故障时,重配置另一条链路的方向来传输数据包.当链路出现细粒度故障时,结合网络通信状态,选择重配置另一条链路的方向或利用自身剩余的完好位线来多次传输数据包.实验结果表明,该文方法较双向链路重配置方案和单向链路多次传输的容错方案而言,延时最多减小27.2%和34.2%,吞吐率最多增大21.7%和35.3%。在链路出现不同情况的故障时,该文方案有着明显的优势,保证了网络的可靠性。     

NoC中负载均衡的AVOQ路由器设计

针对片上网络中使用虚拟输出队列(VOQ)机制的路由器在网络拥塞时存在的头阻塞问题,提出负载均衡的AVOQ路由器架构。首先,输入缓冲区仍使用VOQ机制来处理头阻塞问题。其次,在路由计算模块自适应地选择输出端口,确保数据从较不拥塞的端口输出;在单个虚通道内自适应地读取数据包,确保下游不拥塞的流量能够在网络里流通。实验结果表明,相较于虚通道路由器和VOQ路由器,AVOQ路由器平均延时最多减少83.2%和57.1%,吞吐率最多增加72.7%和33.3%,功耗和面积开销可接受。该方案通过两个层级的自适应均衡全网的流量分布,缓解拥塞,进而降低头阻塞出现的可能性,并在头阻塞出现时消除其影响,提升网络性能。     

WiNoC中拥塞避免的低功耗路由器设计

无线片上网络中无线路由器承担着大量的数据通信任务,容易产生拥塞,影响网络性能。基于此,设计了一种拥塞避免的低功耗路由器。首先,在无线接收缓冲区与路由器的输入缓冲区之间添加数据包分流模块,将无线接收端的大量数据包,存储到空闲端口输入缓冲区;其次,添加全局仲裁模块,授权无线发送端口拥塞度较高的无线路由器,优先使用无线信道;最后,对无通信任务的无线收发器组件进行功率门控。实验结果表明,该方案使得网络的吞吐率最多增加了25%,网络延时最多降低40%,功耗最多降低33.87%,面积开销稍有增加。该设计有针对性地解决了高负载引起的无线收发端拥塞问题,提升了网络整体性能。     

基于“包-电路”交换的双环片上网络设计

针对包交换片上网络(NoC)在大量数据通信情况下性能较差的弱点,提出了一种基于“包-电路”(PCC)交换的环形拓扑结构片上网络(DRNoC)设计架构。首先这种双环形拓扑结构由内外两环构成,可实现环内或环间双向通信,环上节点数目可拓展。其次DRNoC路由器通道可配置为桥节点或环节点路由器两种类型,相比于2D-Mesh型通道数减少,结构更加简单,资源消耗更少。最后提出了针对DRNoC的双环动态路由算法(DDRA),该算法无需在每个路由节点都进行输出方向的译码判断,在头包建立受阻时,根据网络情况选择其他路由路径,最大程度保证数据同环传输基础上跨环传输,有降低头包建立的等待时间,提高吞吐率。实验表明,在大量数据通信情况下,搭载DDRA算法的DRNoC的硬件资源开销降低的同时能够降低网络平均包延时提升平均吞吐率,有效地改善了网络性能。     

面向弹性提升的主动配电网重构与元件修复协同方法

极端事件会导致主动配电网出现多处故障且故障元件修复时间长,传统的故障恢复方法会使得系统的弹性低。针对主动配电网的故障恢复阶段,提出了一种面向弹性提升的重构与故障元件修复协同方法。首先,结合计及负荷重要程度的主动配电网弹性指标,分析出故障恢复阶段弹性提升的主要措施;在此基础上,提出重构与故障元件修复协同策略框架,给出交替协同中线路状态变量和节点负荷供电状态变量的更新与关联。其次,建立全过程故障元件修复和重构的目标函数。针对主动配电网含多分布式电源特点带来的故障元件上下游关系难以判别问题,提出基于消线解故障环法和圈源等效法的判别方法。最后,算例结果表明,所提方法能有效地提升主动配电网的弹性。     

k-NN based fault detection andclassification methods for powertransmission systems

This paper deals with two new methods, based on k-NN algorithm, for fault detection and classification in distance protection. In these methods, by finding the distance between each sample and its fifth nearest neighbor in a predefault window, the fault occurrence time and the faulty phases are determined. The maximum value of the distances in case of detection and classification procedures is compared with pre-defined threshold values. The main advantages of these methods are: simplicity, low calculation burden, acceptable accuracy, and speed. The performance of the proposed scheme is tested on a typical system in MATLAB Simulink. Various possible fault types in different fault resistances, fault inception angles, fault locations, short circuit levels, X/R ratios, source load angles are simulated. In addition, the performance of similar six well-known classification techniques is compared with the proposed classification method using plenty of simulation data     

k-NN based fault detection and classification methods for power transmission systems

This paper deals with two new methods, based on k-NN algorithm, for fault detection and classification in distance protection. In these methods, by finding the distance between each sample and its fifth nearest neighbor in a predefault window, the fault occurrence time and the faulty phases are determined. The maximum value of the distances in case of detection and classification procedures is compared with pre-defined threshold values. The main advantages of these methods are: simplicity, low calculation burden, acceptable accuracy, and speed. The performance of the proposed scheme is tested on a typical system in MATLAB Simulink. Various possible fault types in different fault resistances, fault inception angles, fault locations, short circuit levels, X/R ratios, source load angles are simulated. In addition, the performance of similar six well-known classification techniques is compared with the proposed classification method using plenty of simulation data     

基于状态估计残差比较的配电网故障区段定位方法

针对单端量测的阻抗法在多分支辐射状网络中会出现伪故障点的问题,利用网络中分布的智能量测设备提供的监测信息,提出了基于状态估计残差比较的配电网故障区段定位方法。该方法首先遍历所有节点通过添加故障支路建立不同节点故障网络下的状态方程;其次通过求解状态估计方程获得含网络所有节点的总残差检测值序列;然后搜寻最大残差检测值所在节点判断故障节点位置;最后根据节点相邻区间首末端阻抗角相位变化差异判定故障区间。大量仿真表明,该方法在不同故障类型、过渡电阻等工况下均具有良好的故障定位效果。该方法考虑了量测产生的随机变量成分,利用测量数据的互补性和多源性,在一定程度上提高了不同量测数据误差的抗干扰能力,极大地缩小了配电网在有限量测条件下的故障查找范围。     

A generalized fault diagnosis method in dynamic analogue circuits

Fault diagnosis of analogue circuits is essential for analogue and mixed‐signal systems testing and maintenance. A new method is proposed in this paper for multiple fault diagnosis of linear analogue circuits in frequency domain. The Woodbury formula is applied to the modified nodal equation to construct the fault diagnosis equation, which relates the limited measured circuit responses with the multiple faults inside the circuit in a linear way. A recently developed ambiguity group locating technique is modified here to identify the faulty parameters directly. Computation cost is reduced compared to combinatorial search in traditional fault verification methods. Only one node is needed for voltage measurement, but multiple excitations on accessible nodes are required for fault identification. Parameter evaluation can provide the exact solution to the deviated values of faulty parameters. The faulty parameter deviations can have any finite values. Example circuits are provided to illustrate the proposed method. Two other methods for multiple analogue fault diagnosis sharing the same mechanism as the method proposed in this paper are also briefly described. The proposed method is extremely effective for the circuit with very limited accessible nodes and is also computationally efficient. Copyright © 2002 John Wiley & Sons, Ltd.     

A fault diagnosis method for power systems based on temporal tissue-like P systems

To quickly and accurately identify faulty components based on the alarm information is critical for the fault diagnosis of power grids. To address this challenge, this paper proposes a novel fault diagnosis method based on temporal tissue-like P system (TTPS). In the proposed method, suspected faulty components are identified first via a network topology analysis method. An TTPS-based fault diagnosis model is then built for each suspected faulty component to perform fault reasoning, so as to accurately detect the faulty components. To take full advantage of the action signals and temporal information of protection devices, TTPS and its forward temporal reasoning algorithm are proposed. TTPS can synchronously model the action and temporal logics of protection devices in an intuitive and graphical way, while the reasoning algorithm can process the fault alarm information in parallel. To demonstrate the effectiveness and superiority of the proposed method, simulations are carried out on the IEEE 14-bus and 118-bus systems, while the results are compared to other two widely adopted methods.     

含光伏电源的配电网故障定位策略研究

基于监测电压暂降的配电网故障定位逐渐成为研究热点,同时随着智能配电网的发展,光伏电源越来越多的接入配电网。光伏电源的接入使传统的辐射型配电网的网络结构发生变化,且光伏电源的故障电流特性与同步发电机不同,使得基于电压暂降的配电网故障定位方法运用于含光伏电源的配电网时会产生较大误差。为了解决含光伏电源配电网的故障定位问题,基于低电压穿越控制策略针对光伏电源的短路电流特性进行了分析,然后介绍了含光伏电源的配电网故障电流的求解算法。通过遍历搜索候选母线,确定使得电气故障后故障电压实测值与理论计算值最为吻合的母线和故障类型,从而确定故障最邻近的母线。基于对IEEE34节点配电系统改造后的算例进行计算仿真,验证了文章方法应用于含光伏电源配电网定位的正确性和有效性。     

A new faulty section identification and fault localization technique for three-terminal transmission line

Owing to mal-operation of the conventional scheme during high resistance ground fault near tap point, a new faulty section identification and fault localization technique for three-terminal transmission line is presented in this paper. The proposed technique utilizes time-synchronized voltage and current signals from all the three terminals. Initially, fault detection based on estimation of superimposed voltage of tap point with reference to all three terminals has been carried out. Subsequently, utilizing the above three estimated superimposed voltages; faulty section identification criterion is formed. Finally, fault localization i.e. estimation of the value of fault distance and resistance has been performed. The authenticity of the proposed technique has been verified by simulating an existing 400 kV Indian three-terminal transmission network in PSCAD/EMTDC software. The simulation results point out that the proposed technique is able to identify the faulty section correctly. Moreover, it precisely estimates the value of fault distance and fault resistance as the percentage error for fault location and resistance remains within ±1.5% and ±3.5%, respectively. Likewise, its performance remains unaffected during wide variation in fault and system parameters. At the end, comparative evaluation of the proposed technique with the existing protection scheme clearly shows its superiority.     

A single ended directional fault section identifier and fault locator for double circuit transmission lines using combined wavelet and ANN approach

Fault section identification and determining its location are important aspects to reduce down/repair time, speed up restoration of power supply and to improve the reliability. In this paper combined wavelet and artificial neural network based directional protection scheme is proposed for double circuit transmission lines using single end data to identify the faulty section and its location with reach setting up to 99% of line length. The proposed method requires the three phase currents and voltage to be measured at one end of the double circuit transmission line modelled using distributed parameter line model which also considers the effect of shunt capacitance. Approximate coefficients feature vector of the three phase voltage and current are extracted using discrete wavelet transform to train the ANN with Levenberg Marquardt algorithm. The proposed scheme involves two stages. The first stage identifies the zone/section of the fault and the second stage calculates the fault location from the relaying point. The proposed combined Wavelet and ANN based fault location scheme is also compared with ANFIS based fault location scheme. The test results of the proposed scheme show that the fault section identification and location estimation is very accurate and the average percentage error in fault location estimation is within 0.001%. This method is adaptive to the variation of fault type (both forward and reverse), fault inception angle, fault location and fault resistance. The main advantage of the proposed scheme is that it offers primary protection to 99% of line length using single end data only and also backup protection to the adjacent forward and reverse line sections.