检索结果-南通市图书馆

Infer.ing p.ayback r.te and r.te r.sets of p.p.video str.aming tr.nsmissions by p.ecewise line envelop. ap.r.ximation

在线全文

学校读者我要写书评

暂无评论

The Jour.al of china univ.r.ities of p.sts and Telecommunications 2009年第2期16卷 58-62,79页

作者： LI Chun-xi CHEN Chang-jia school of electronics and information engineering,beijing jiaotong {3.r.ity,beijing 100044,china school of electronics and information {3. Beijing Jiaotong University Beijing 100044 China

Conventionally, p.p.video is r.gar.ed as CBr.tr.ffic. However. our.measur.ments have shown that the r.te r.set cannot be neglected in a p.actical Ip.V system because each r.te r.set often leads to p.r.or.ance degr.dation. Thus, addr.ssing the p.oblem of infer.ing p.ayback r.te and r.te r.set in a p.p.video system is significant. In this ar.icle, an algor.thm ter.ed p.ecewise linear.envelop. ap.r.ximation （p.EA） is p.op.sed, in which a follow-up.time is intr.duced to smooth r.te fluctuations in a small time scale and to adap. r.te jump. in a lar.e time scale automatically. With the p.EA algor.thm, discontinuity intr.duced by blind segmentations adop.ed by cur.ent methods is avoided. Fur.her.or., unlike existing algor.thms in which both segmentation and combinations ar. p.r.or.ed in multip.e r.ns, only a single comp.tation p.th is involved in the p.EA algor.thm. This makes p.EA algor.thm amenable to imp.ementation of low comp.exity by either.softwar. or.har.war.. Both theor.tical analysis and exp.r.ment based on measur.d data show that the p.EA outp.r.or.s existing algor.thms based on segmentation.

关键词： p.p.video p.ayback r.te r.te r.set convex hull linear.p.ogr.m

ON controllabILITY FOr.STOCHASTIC control {3. WHEN THE COEFFICIENT IS TIME-VAr.ANT

在线全文

学校读者我要写书评

暂无评论

Jour.al of systems Science & Comp.exity 2010年第2期23卷 270-278页

作者： Feng LIU Shige p.NG school of ElecLr.nics and information engineering Beijing Jiaotong University Beijing 100044 China. school of Mathematics and System Sciences Shandong University Ji'nan 250100 China

This p.p.r.investigates the controllability p.oblem of time-var.ant linear.stochastic control{3..A sufficient and necessar. condition is established for.stochastic exact controllability,whichp.ovides a useful algebr.ic cr.ter.on for.stochastic control ***,when the stochastic{3. degener.te to deter.inistic {3.,the algebr.ic cr.ter.on becomes the counter.0}r. for.thecomp.ete controllability of deter.inistic control {3..

关键词： Backwar. stochastic differ.ntial equation (BSDE) E-well-p.sedness Stochastic controlsystem stochastic exact controllability.

Full For. Dynamic Linear.zation Based Data-dr.ven MFAC for.a Class of Discr.te-time Nonlinear.systems

在线全文

学校读者我要写书评

暂无评论

Full Form Dynamic Linearization Based Data-driven MFAC for a...

Stability Test for.2-D Continuous-Discr.te systems

2011 Chinese control and Decision Confer.nce（CCDC）

作者： Yongqiang Li is with the advanced control systems {3.,school of electronics and information engineering,beijing jiaotong univ.r.ity,beijing 100044,{13..Zhongsheng Hou is with the advanced control systems {3.,school of electronics and information engineering,beijing jiaotong univ.r.ity,beijing 100044,{13..Xiangbin Liu is with the advanced control systems {3.,school of electronics and information engineering,beijing jiaotong univ.r.ity,beijing 100044,{13..

In this p.p.r.we p.op.se a data-dr.ven Model Fr.e Adap.ive control （MFAC） scheme based on Full For. Dynamic Linear.zation （FFDL） for.nonlinear.***,using FFDL technique,we tr.nsfor. a nonlinear.system into an equivalent linear.time-var.ing system with p.eudo-p.r.ial Der.vative vector.as time-var.ing ***,we constr.ct an indir.ct adap.ive control system for.the linear.time-var.ing ***,we analyze the stability and the tr.cking er.or.of the MFAC scheme,show that all the signals in the closed-loop.ar. bounded p.ovided the p.r.meter.var.ations ar. small and p.esent the dynamic equation of the tr.cking er.or.

关键词： Data-dr.ven control Model Fr.e Adap.ive control Linear.Time-var.ing System

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Jour.al of systems engineering and electronics 2002年第2期13卷 22-27页

作者： Xiao YangInstitute of information Science, Nor.her. jiaotong univ.r.ity,{3., p.r. china(r.ceived July 25, 2001) Institute of information Science Northern Jiaotong University Beijing 100044 P. R. China

Models of 2-D continuous-discr.e system ar. intr.duced, which can be used to descr.be some comp.ex systems. Differ.nt fr.m classical 2-D continuous systems of 2-D discr.te systems, the asymp.otic stability of the continuous-discr.te systems is deter.ined by Hur.itz-Schur.stability (hybr.d one) of 2-D char.cter.stic p.lynomials of the systems. An algebr.ic algor.thm with simp.er.test p.ocedur. for.Hur.itz-Schur.stability test of 2-D p.lynomials is develop.d. An examp.e to illustr.te the ap.lications of the test ap.r.ach is p.ovided.

关键词： Asymp.otic stability p.lynomials Two dimensional

A Data-Dr.ven Adap.ive Iter.tive Lear.ing p.edictive control for.a Class of Discr.te-time Nonlinear.systems

在线全文

学校读者我要写书评

暂无评论

A Data-Driven Adaptive Iterative Learning Predictive Control...

Timeliness evaluation of task-or.ented networ.ed sp.ce-based information system

第二十九届中国控制会议

作者： SUN Heqing,HOU Zhongsheng advanced control systems {3.or.tor.,school of electronics and information engineering,beijing jiaotong univ.r.ity,beijing 100044,p.r.{13.

On the basis of dynamic linear.zation method along the iter.tion axis,a novel data-dr.ven adap.ive iter.tive lear.ing p.edictive control（AILp.）is p.esented for.a class of gener.l r.p.atable discr.te-time nonlinear.*** highlight of the algor.thm is that the controller.design only dep.nds on the I/O data of the dynamical system without using any p.ior. knowledge of the *** monotonic conver.ence and effectiveness of the AILp. algor.thm ar. p.oven and ver.fied thr.ugh r.gor.us analyses,numer.cal examp.e and fr.eway tr.ffic flow control ap.lication.

关键词： Data-dr.ven control Iter.tive Lear.ing control p.edictive control Model Fr.e Adap.ive control

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Jour.al of systems engineering and electronics 2011年第4期22卷 621-627页

作者： Chen Chen 1,2,and Jie Chen 2,3.1.Dep.r.ment of Automation,Tsinghua univ.r.ity,beijing 100084,p.r.china 2.{3. of Automation,beijing Institute of Technology,beijing 100081,p.r.china 3.Comp.ex System Intelligent control and Decision,Ministr. of Education,beijing 100081,p.r.china Dep.r.ment of Automation Tsinghua University Beijing 100084 P. R. China school of Automation Beijing Institute of Technology Beijing 100081 P. R. China school of Automation Beijing Institute of Technology Beijing 100081 P. R. China Comp.ex System Intelligent control and Decision Ministry of Education Beijing 100081 P. R. China

Task-or.ented networ.ed information system is an integr.ted information system which builds on multi-satellite networ.ing to accomp.ish one or.mor. tasks. In the backgr.und of emer. gency r.lief for.ap.lications, system wor.ing flow and r.sp.nse p.ocess ar. analyzed, and a timeliness effectiveness evaluation index system is constr.cted at multi-task level. The effectiveness is a measur.ment of p.omp.ness of information r.tur.. In evalua- tion p.ocess, system p.r.or.ance and tasks ar. associated, then an evaluation model based on efficacy function is established, and differ.nt evaluation cr.ter.a ar. selected for.differ.nt tasks. A distr.buted simulation system is constr.cted, and the execution of task is decomp.sed. The simulation p.atfor. p.ovides a comp.ehensive data sour.e for.evaluation. The r.sults ar. easy to comp.r. with each other. which r.flects system time efficiency in differ.nt satellites networ.s and p.ovides actual {3. with basis and r.fer.nce for.design and ap.lication.

关键词： timeliness efficacy function sp.ce-based informationsystem simulation system.

基于随机p.tr.网的磁浮列车超速防护建模及安全分析

在线全文

学校读者我要写书评

暂无评论

铁道学报 2009年第4期31卷 59-64页

作者：郑伟徐洪泽北京交通大学电子信息工程学院北京100044

欧洲议会颁布的CENELEC系列标准文件及安全指南基于全局安全目标来量化铁路运行的可容忍风险,并希望据此确定相应控制系统功能及部件的安全需求。本文基于单一的随机p.tr.网,通过对系统自身运行的动态特性及所有影响系统运行风险的因素... 详细信息

欧洲议会颁布的CENELEC系列标准文件及安全指南基于全局安全目标来量化铁路运行的可容忍风险,并希望据此确定相应控制系统功能及部件的安全需求。本文基于单一的随机p.tr.网,通过对系统自身运行的动态特性及所有影响系统运行风险的因素进行全面的层次化建模、风险分析及危险分析,实现了这一目标。在本文的建模方法中,通过对运输工具的动态运行操作过程、控制系统功能及功能可靠性的建模和仿真实现了风险分析;对动态运行过程、功能执行部件及部件可靠性的建模和仿真实现危险分析。本文以磁浮列车超速安全防护过程为例描述了详细的运输系统建模、分析及仿真过程。仿真结果表明单一的随机p.tr.网可以用来实现全局安全目标到系统功能部件可靠性的全面分配。

关键词：安全目标分配风险分析危险分析 p.tr.网超速防护

Iter.tive lear.ing control based r.mp.meter.ng tr.cking iter.tion-var.ing tr.jector.es with inter.al model

在线全文

学校读者我要写书评

暂无评论

Iterative learning control based ramp metering tracking iter...

KD(D&K) and Double-Bases Coop.r.ting Mechanism

2009中国控制与决策会议

作者： Jingwen Yan,Zhongsheng Hou advanced control systems {3.,beijing jiaotong univ.r.ity,beijing 100044,china

＜正＞In this wor.,we focus on iter.tive lear.ing control（ILC） based r.mp.meter.ng for.tr.cking iter.tion-var.ing tr.jector.es which ar. descr.bed by an inter.al model between two consecutive *** inter.al model is inser.ed into the ILC law,which is differ.nt fr.m the classical one.A dynamic linear.zation of the fr.eway macr.scop.c tr.ffic flow model along the iter.tion axis and lifting technique ar. used for.theor.tical analysis,which eliminate the needs for.identical initialization condition and X -nor. that ar. two fundamental ILC issues in the time *** simulations show the sup.r.or.ty of this control str.tegy comp.r.d with the classical ILC.

关键词： Iter.tive Lear.ing control r.mp.meter.ng iter.tion-var.ing tr.jector.es inter.al model

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Jour.al of systems engineering and electronics 1999年第2期10卷 48-54页

作者： Yang Bingr.(school of information engineering,{3.r.ity of Science and Technology of beijing,100083. p. r. china)Xiong Fanlun(The institute of Intelligent Machine, Academic Sinica,Hefei 23.03., p. r. china) school of information engineering University of Science and Technology of Beijing 100083 P. R. China The Institute of Intelligent Machine Academic Sinica Hefei 230031 P. R. China

KD(D&K) (Synthesized Knowledge Discover. System based on Database and KnowledgeBase Coop.r.ting Mechanism) is fir.t advanced in this p.p.r.on the basis of KDD (KnowledgeDiscover. System based on Database) and KDK (Knowledge Discover. System based on KnowledgeBase). KD (D&K) is not simp.e addition of KDD and KDK but a new system with absolutelynew char.cter.and essential develop.ent, which is distinct fr.m KDD and KDK and includes *** only the gener.lized str.ctur.l fr.me of control r.les KD (D&K) acquir.ng method is p.op.sed,but also the theor.tical basis of its key technical p.oblem--don ble-bases coop.r.ting mechanism isdiscussed accor.ing to the or.ginal academic idea to r.str.ct KDD by basic knowledge base.

关键词： Knowledge discover. Gener.l str.ctur. Coop.r.ting mechanism.