An Efficient Priority-Driven Congestion Control Algorithm for Data Center Networks

With the emerging diverse applications in data centers,the demands on quality of service in data centers also become diverse,such as high throughput of elephant flows and low latency of deadline-sensitive flows.However,traditional TCPs are ill-suited to such situations and always result in the inefficiency(e.g.missing the flow deadline,inevitable throughput collapse)of data transfers.This further degrades the user-perceived quality of service(QoS)in data centers.To reduce the flow completion time of mice and deadline-sensitive flows along with promoting the throughput of elephant flows,an efficient and deadline-aware priority-driven congestion control(PCC)protocol,which grants mice and deadline-sensitive flows the highest priority,is proposed in this paper.Specifically,PCC computes the priority of different flows according to the size of transmitted data,the remaining data volume,and the flows’deadline.Then PCC adjusts the congestion window according to the flow priority and the degree of network congestion.Furthermore,switches in data centers control the input/output of packets based on the flow priority and the queue length.Different from existing TCPs,to speed up the data transfers of mice and deadline-sensitive flows,PCC provides an effective method to compute and encode the flow priority explicitly.According to the flow priority,switches can manage packets efficiently and ensure the data transfers of high priority flows through a weighted priority scheduling with minor modification.The experimental results prove that PCC can improve the data transfer performance of mice and deadline-sensitive flows while guaranting the throughput of elephant flows.

A Hybrid Model Predictive Control for Handling Infeasibility and Constraint Prioritization

A hybrid approach using MLD (mixed logical dynamical) framework to handle infeasibility and constraint prioritization issues in MPC (model predictive control) based on input-output model is introduced. By expressing constraint priorities as propositional logics and by transforming the propositional logics into inequalities,the infeasibility and constraint prioritization issues are solved in the MPC. Constraints with higher priorities are met first, and then these with lower priorities are satisfied as much as possible. This new approach is illustrated in the control of a heavy oil fractionator-Shell column. The overall control performance has been significantly improved through the infeasibility and control priorities handling.

Indoor biology pollution control based on system-based humidity priority control strategy

Indoor biological contamination and HVAC system secondary contamination problems caused wide public concerns. Biological contamination control will be the next step to achieve better IAQ. The most efficient and safe way to control biological contamination was to limit relative humidity in HVAC system and conditioned environment in the range that is more unsuitable for microorganism to survive. In this paper,by referring to bio-clean project experiences,a system-based humidity priority control manner came into being by lowering outdoor air humidity ratio to eliminate all indoor latent load and using self recirculation units to bear indoor sensible load. Based on the whole-course residue humidity contaminant control concept,dynamic step models for coil and conditioned zone were developed to describe mass and energy conservation and transformation processes. Then,HVAC system and conditioned zone dynamic models were established on LabVIEW+Matlab platform to investigate optimized regulation types,input signatures and control logics. Decoupling between cooling and dehumidification processes can be achieved and a more simplified and stable control system can be acquired by the system-based humidity priority control strategy. Therefore,it was a promising way for controlling biological pollution in buildings in order to achieve better IAQ.

An ATM Switch Architecture with Priority Control

A new ATM switch architecture based on the Knockout switch is proposed.Prioritycontrol function is assigned to the hardware of switch to meet the requirements of cell lossprobabilities and delay characteristics of different services.This architecture can use systemresource more effectively,meet the QOS(quality of service)requirements of different cus-tomers,and also can reduce the complexity of switch.

Bus Priority Control for Dynamic Exclusive Bus Lane

One problem with the existing dynamic exclusive bus lane strategies is that bus signal priority strategies with multi-phase priority request at intersections are not adequately considered.The principle of bus signal priority level was designed based on the isolated multi-phase structure principle consideration of the bus signal priority,and a new priority approach for the dynamic exclusive bus lane was proposed.Two types of priority strategies,green extension and red truncation,were proposed for current phase and next phase buses,respectively.The control parameters including minimum green time,green extension time,maximum green time and bus arrival time are calculated.The case studies for this paper were carried out using four consecutive intersections of Huaide Middle Road in Changzhou City.The signal control scheme was designed using the conventional,exclusive bus lane method,the dynamic exclusive bus lane without signal priority method,and the proposed approach,respectively.The authors used the VISSIM simulation platform to evaluate the efficiency of each approach.Results showed that the method of approach can significantly decrease delays caused by social and conventional buses and make up for the negative impact social buses have on the bus rapid transit(BRT)operation,which allows the method to complement the dynamic,exclusive bus lane design.

Control strategies and hardware-in-the-loop simulation for bus signal priority

It is considered as an important and effective means to give priority to the development of public transit which can improve the efficiency of transportation resources utilization and alleviate traffic jams. Public transit signal priority belongs to the "time priority" among the right-of-way priorities. After reviewing the existing bus priority signal control strategies and the advances in related technologies at home and abroad, this article analyzed the breakthrough direction of the bus signal priority design technologies suitable for China's conditions, and then proposed the hardware and software systems and the modules for the bus priority signal control system. Finally, the hardware-in-the-loop simulation (HILS) was introduced to evaluate bus priority signal control programs in order to optimize the control strategies.

Boundedness and liveness enforcement for labeled Petri nets using transition priority

This paper deals with the supervisory control problem of discrete event systems modeled by labeled Petri nets. The system is originally unbounded. First, the solvability of the problem is confirmed. A necessary condition is given and proven for the existence of a feasible priority-based controller based on the notions of liveness and transition invariants. Next, a cyclic behavior graph is constructed, which shows the reachable markings that guarantee the maximum liveness of the system within a given bound vector. Finally, an on-line control strategy is proposed to enforce boundedness and liveness to the given system by appending priority relations to transitions. The dynamic priority relation changes flexibly according to the current state of the system and enforces the system evolving in a bounded and live manner. In addition, numerical examples are studied to verify the validity of the proposed approach that remains the structure of the plant net and is efficient for on-line control.

Temporal Preferences-Based Utility Control for Smart Homes

The residential sector contributes a large part of the energy to the global energy balance.To date,housing demand has mostly been uncontrollable and inelastic to grid conditions.Analyzing the performance of a home energy manage-ment system requires the creation of various profiles of real-world residential demand,as residential demand is complex and includes multiple factors such as occupancy,climate,user preferences,and appliance types.Average Peak Ratio(A2P)is one of the most important parameters when managing an efficient and cost-effective energy system.At the household level,the larger relative magni-tudes of certain energy devices make managing this ratio critical,albeit difficult.Various Demand Response(DR)and Demand Side Management(DSM)systems have been proposed to reduce this ratio to 1.The main ways to achieve this are economic incentives,user comfort modeling and control,or preference-based.In this study,we propose a unique opportunistic social time approach called the Time Utility Based Control Feature(TUBCF),which uses the concept of a utility function from economics to model and control consumer devices.We propose a DR model for residential customers to reduce Peak-to-Average Ratio(PAR)and improve customer satisfaction by eliminating Appliance Wait Time(WTA)during peak periods.For PAR reduction and WTA,we propose a system architecture and mathematical formulation.Our proposed model automatically schedules devices based on their temporal preferences and considers six households with different device types and operational characteristics.Simulation results show that using this strategy can reduce A2P by 80%and improve user comfort during peak hours.

An IEEE 802.3 Compatible Real Time Medium Access Control with Length-based Priority

A new medium access control method is proposed over the predominant Ethernet broadcast channel. Taking advantages of intrinsic variable length characteristic of standard Ethernet frame, message-oriented dynamic priority mechanism is established. Prioritized medium access control operates under a so-called block mode in event of collisions. High priority messages have a chance to preempt block status incurred by low priority ones. By this means, the new MAC provides a conditional deterministic real time performance beyond a statistical one. Experiments demonstrate effectiveness and attractiveness of the proposed scheme. Moreover, this new MAC is completely compatible with IEEE802.3.

基于公交优先的干线协调信号控制改进模型

基于公交信号优先策略,对MULTIBAND模型进行改进,重新定义了权重系数、车道清空时间和最小绿波带宽,结合不同相位放行方式实施红灯早断或绿灯延长控制策略,并在此基础上考虑公交车辆的车速与停靠时间,建立对公交车辆的约束,提出基于公交优先的干线协调信号控制改进模型。通过VISSIM仿真,将该模型与传统干线协调信号控制模型对比分析,结果表明:与传统协调控制模型相比,模型使社会车辆平均延误降低了4.77%,停车次数降低了3.77%,公交车辆平均延误降低了11.35%,停车次数降低了2.06%,人均延误降低了7.22%,且在不同的公交流量下均有较好的效果。

卡片导航平衡控制系统性能仿真研究

为了减少装配作业车间内因物料齐套产生的等待浪费,使具有装配约束的关联零件加工进度得到有效协同,设计一类卡片导航平衡控制系统(control of balance by card-based navigation,COBACABANA)。其基于两类卡片循环回路实现任务投放与作业分派的可视化进度协同控制逻辑。本文详细介绍系统的运行机制和系统控制参量,通过构建一般化的装配作业车间仿真模型,探讨在不同装配关联度下各控制参量的性能变化。实验结果表明,COBACABANA系统性能良好,并且选择合适的控制参量就能够有效提升关联零件的进度协同性。

中国潜在自主可控关键产品识别与突破路径研究

系统分析中国潜在自主可控关键产品突破时序路径,是逆全球化背景下全面防范和化解“卡脖子”潜在风险的关键。基于1995—2021年243个国家(地区)出口产品数据构建产品空间网络模型,识别全球核心产品及中国潜在自主可控关键产品,基于产品能力和网络核心性构建优先度指数,挖掘潜在自主可控关键产品突破路径。研究表明,在2021年核心度前10%的124种全球核心产品中,共有54种中国潜在自主可控关键产品,其突破路径为:充分利用机械与电力设备、基本金属制品、化工产品等领域已有的产品能力基础,优先突破光学显影和金属工艺等领域的基础关键产品,再进一步突破机械与电力设备、光学与医疗等精密仪器、塑料与橡胶制品业等领域高端关键产品。

巴伊盆地地下水优先控制污染物筛选

为确定巴里坤-伊吾盆地地下水优先控制污染物,文章采用盒须图法和累积频率法对地下水中无机指标的背景值进行厘定,使用以分类组合为基础的改进内梅罗指数法筛选地下水无机优先控制污染物;采用基于检出率、检出浓度和污染物毒理学性质构建的有机优先控制污染物识别方法对有机污染物进行筛选;采用绝对主成分分析-多元线性回归模型和土地利用类型分析优先控制污染物来源。结果表明:无机优先控制污染物有TH、TDS、Na^(+)、Cl^(-)、SO_(4)^(2-)、NO_(3)^(-)、Fe和F^(-),其中,TH、TDS、Na~+、SO_(4)^(2-)和Cl^(-)主要受到与水-岩相互作用有关的自然影响因素,F^(-)和NO_(3)^(-)主要来源于农业生产活动,Fe主要来源于矿业、工业活动;有机优先控制污染物为1,2-二氯苯,来源于农业生产中化肥农药的使用。

基于MEC服务器优先服务的路侧单元MAC层调度策略

针对多接入边缘计算(MEC)服务器高可靠、低时延和大数据量的数据传输要求,基于无冲突接入、优先级架构和弹性服务技术,提出一种适用于车辆边缘计算场景下的媒体访问控制(MAC)调度策略。所提策略由车联网(IoV)路侧单元(RSU)集中协调信道接入权,优先确保车载网络中车载通信单元(OBU)与MEC服务器之间的链路传输质量,以及时传输车辆到网络(V2N)业务数据;同时,对本地OBU之间的业务采取弹性服务方式,增强密集车辆接入时应急消息传输的可靠性。首先,构建调度策略的排队分析模型;其次,根据各时刻系统状态变量的无后效性特点建立嵌入式马尔可夫链,并通过概率母函数的分析方法对系统进行理论分析,得到MEC服务器通信单元和OBU的平均排队队长、平均等待时延和RSU查询周期等关键指标的精确解析表达式。计算机仿真实验结果表明,统计分析结果与理论计算结果一致,所提调度策略在高负载情况下能够提高IoV的稳定性和灵活性。

基于直接磁链跟踪控制的无刷双馈电机无Crowbar对称低电压穿越方法

低电压穿越(low-voltage ride through,LVRT)能力是并网型风电机组重要的性能指标之一。该文提出一种基于直接磁链跟踪控制的无刷双馈感应发电机(brushlessdoubly fed induction generator,BDFIG)无Crowbar对称低电压穿越方法。该方法基于开关表,通过直接选定电压空间矢量来控制无刷双馈电机控制绕组(control winding,CW)磁链的幅值和相位,以迅速抑制低电压穿越期间控制绕组的感应过电压和由此导致的过电流,同时快速向电网注入动态无功电流以支撑电压恢复。该方法无需坐标变换、系统结构简单、磁链动态响应迅速。样机实验结果表明,低电压穿越过程中,有功恢复及无功功率动态响应满足国标相关要求,且控制绕组电流得到有效抑制,验证了所提控制策略的有效性。

无人机自组网中基于优先级和发送概率的低时延MAC协议设计

媒体接入控制(MAC)协议是满足数据传输质量服务需求的一项关键技术,其性能的优劣直接决定了网络的整体运行效率。相较于其他经典MAC协议,基于统计优先级的多址接入(SPMA)协议采用了多优先级接入、信道检测和流量控制等机制,因此它能更好地满足无人机自组网的低时延、区分服务、大规模组网等重要需求。其中,不同优先级阈值是影响该协议性能的关键指标之一,而现有研究中的阈值设置策略会导致无人机自组网吞吐量波动较大的问题。除此之外,在大规模节点组网的情况下,由于特殊的发送机制,该协议面临数据包冲突加剧的问题。针对以上问题,本文提出了一种新的阈值设置方法和低优先级数据概率性接入信道的方法。在大规模节点网络高业务量的情况下,该改进策略根据新的阈值和周期性负载统计的比较结果决定是否延迟或者拒绝相关数据的接入。与此同时,将SPMA协议的多优先级接入机制和排队论基本原理相结合,通过计算设定发送概率对不同低优先级数据进行接入控制。仿真结果表明,所提方法能够使得无人机自组网的吞吐量保持稳定,并为不同类型业务的差异性服务质量(QoS)需求提供良好支持,提高不同优先级数据的首发传输成功率的同时不会带来较大的额外时延损耗,性能优于现有的SPMA协议。

基于子网优先级驱动的交直流混合微网集群双向互联变流器分散式控制策略

随着可再生能源渗透率的不断提高,交直流混合微电网集群逐渐成为新型电力系统的弹性解决方案。考虑到实际运行中各子电网优先级和电能质量需求存在差异,该文提出一种基于子网优先级驱动的交直流混合微网集群双向互联变流器分散式控制策略,以实现系统的弹性功率交互。首先,通过将归一化的交流频率和直流电压值作为子电网的状态偏差因子,以最小化全局总偏差因子为目标构建了用于交直流混合微电网集群的最优电力交换模型,该模型充分考虑了子电网的优先级指数和供电容量;其次,基于最优电力交换目标,提出用于双向互联变流器的准下垂控制策略,由于该策略中双向互联变流器仅需检测本地交流频率/直流电压的状态偏差因子,因此可以实现对交直流混合微电网集群整体偏差因子最小化的分散式最优功率自主协同控制;最后,构造了两个仿真实例,并利用半实物仿真系统验证了所提策略的有效性。

基于ControlLogix控制器的双电梯调度控制系统设计

针对双电梯系统的调度问题,设计并实现了一种双电梯调度控制系统。采用罗克韦尔自动化公司的ControlLogix系列可编程逻辑控制器以及PowerFlex40变频器,通过模糊控制方式实现基于呼梯信号优先度的电梯控制方法,利用FuzzyLogix toolbox开发了一套基于交流变频调速和可编程逻辑控制器的双电梯并联调度系统。仿真结果表明,该调度系统达到了距离较近先响应、长时等待先响应的要求。该应用考虑了人性化和实用化等因素,满足了设计使用需求。

西北某工业园区地下水优先控制污染物的识别及风险评估

为了探析西北某工业园区地下水的污染现状,确定地下水中优先控制污染物,了解暴露于污染物所致的健康风险。基于地下水污染物监测结果,利用潜在危害指数、综合评价法识别地下水中优控污染物,使用健康风险评价模型评价暴露于优控污染物的健康风险。结果表明:地下水中氨氮、锰、铅、镉、氯化物、三氯甲烷等污染物超标,污染指数范围为0.005~14.349。地下水中优控污染物包括二氯甲烷、铅、砷、汞等10种污染物。儿童和成人暴露于污染物的累积非致癌风险较小,经饮水摄入和皮肤接触暴露于三氯甲烷和汞有致癌风险,经饮水暴露砷有致癌风险;成年人经皮肤接触暴露砷有致癌风险。工业园区地下水部分污染物超标现象突出,应重点关注优控污染物,优控污染物对儿童和成人有不同程度的健康风险。