摘要
论文题目：交叉口群拥堵扩散机理及其控制与诱导协同模型研究
随着我国城市经济的飞速发展、城市化的不断加快、居民生活水平的逐步提高，人们对交通的需求也越来越高，促使交通运输与社会经济发展和人民日常生活更加紧密。这种现象在大中城市当中更加明显，居民收入的不断增加促使机动车的保有量激增，当相应的交通管理无法同步时，就会导致拥堵、事故等交通问题越来越多，间接引起的能源浪费、环境污染、经济损失等问题也成为影响城市可持续和谐发展的不利因素。其中，交通拥堵问题尤为严重，不仅增加了城市居民的出行时间，降低了居民的出行效率，而且给居民出行带来不便，进而大大降低了城市交通系统的有效运行。
在城市路网中，道路的通行能力包括两方面：路段通行能力和交叉口通行能力。而信号控制交叉口作为间断交通流设施，通过人为的设置红灯和绿灯交替信号，在方便不同转向的车辆快速通过交叉口的同时，也会阻断交通流的连续性，导致交叉口的通行能力要低于路段通行能力，因此交叉口往往成为交通拥堵的诱发点。当关键交叉口出现拥堵时，如果不及时采取有效措施，拥挤会迅速向上游与之关联的交叉口扩散，容易造成区域交通拥挤。因此，解决好由关键交叉口及其关联性密切的相邻节点形成的交通拥堵问题，是目前交通领域研究的重点。
本论文在总结国内外交叉口群研究的基础上，深入剖析交叉口群的交通拥堵扩散机理及其对通行能力的影响分析。利用复杂网络研究城市道路网络的复杂性，同时基于复杂网络群落结构特征提出交叉口群的构想。基于交叉口群中各交叉口这一间断交通流设施的交通流特性，应用“交通波”理论及不同排队长度对上游路段交通流的阻滞效应，同时考虑车辆的实际运行情况，分析排队长度对干道及交叉口通行能力的影响规律，研究了交叉口群的容量特性，为下一步交叉口群的交通状态判别及信号控制与诱导系统的协同研究提供理论基础。
论文考虑交叉口群的交通状态在交通信号控制和动态交通诱导协同中的重要作用，以交叉口群的交通状态判别为研究对象。首先研究了交通状态判别的基础部分交通参数预测，在分析短时交通流的特点的基础上，细致分析了Elman神经网络的特点，提出了基于可变增益因子的动态递归神经网络路段平均速度短时预测方法，利用实时数据对方法的有效性进行了验证。在获取准确的未来交通参数数据基础上，提出了新的交叉口群基本单元的想法，即将关键交叉口和相关路段看成一个基本单元，利用交叉口饱和度、路段平均速度和时间占有率三个参数对基本单元的交通状态进行组合判别，利用长春市人民大街和自由大路交叉口及其南进口路段的实测数据对模型进行了验证，结果表明组合判别方法可以提高判别精度，减少单一参数带来的误差。同时，还给出交叉口过饱和状态的排队长度计算公式，可以为行程时间预测提供准确的路段长度，方便交通诱导时动态交通诱导信息的计算。
交叉口群的实时信号控制及动态交通诱导协同模型是本论文研究的重点，本文综合考虑信号控制和交通诱导的关系以及两者对交通流的影响，对交叉口群的两者协同的可行性进行了分析，利用前景理论对无诱导信息的交通出行、通过可变信息板或交通广播获取诱导信息的交通出行以及通过车载终端获取备选路径信息的交通出行下的驾驶员路径选择行为进行研究，给出三种情形下的驾驶员路径选择概率，并以此为基础建立了准系统最优的动态交通分配模型，最终建立了交叉口群信号控制与动态诱导的双层协同优化模型，同时利用仿真路网数据对模型的有效性进行了验证。
在第四章双层协同优化模型研究的基础上，论文第五章综合考虑驾驶员在出行过程中的信息需求以及驾驶员对诱导信息的选择行为，针对交叉口群的特点，从调查数据的实际出发对交叉口群这种特殊对象下的信号控制与交通诱导协同信息的发布内容、信息发布方式、救援车辆的协同信息优化等方面进行了研究，从而实现事故发生后救援车辆可以快速到达救援现场从而实现紧急救援的目的，最后利用仿真方法对事故条件下的救援车辆与社会车辆的协同管理进行了验证。
With the rapid development of urban economy, the accelerating of urbanization andresidents’ living standard improved gradually in China, the demand for transportation isbecoming much more, causes the relationship between transportation and economydevelopment with daily life more closely. This phenomenon is more obvious in large andmedium-sized cities, residents’ increasing income stimulates the growth of vehicles,when the corresponding traffic managements can’t synchronize, it will lead to more andmore traffic problems such as congestion, accidents and indirectly cause energy waste,environmental pollution, economic loss problems, it also become a negative factor ofinfluencing urban sustainable and harmonious development. Among these problems,traffic congestion is particularly serious, which not only increases residents’ travel time,reduces travel efficiency, but also causes inconvenience to residents’ travel, whichgreatly reduces the effective operation of the urban traffic system.
In urban road network, traffic capacity includes two aspects: road traffic capacityand intersection traffic capacity. As interrupted traffic flow facilities, the signalintersection always set alternate red and green light signal artificially, these can bringvehicles convenience in passing the intersection for different directions, at the same time,it might block the continuity of traffic flow, lead to the traffic capacity of intersectionless than section capacity. Therefore intersections are seen as trigger points of trafficcongestion. When the key intersection appears congestion, if not take timely andeffective measures, the congestion will be quickly transformed to upstream associatedintersection, likely cause regional traffic congestion. Therefore, solving congestionproblem of intersections group formed by the key intersections and associated closelyadjacent nodes is the focus research in transportation field.
Basing on the study of the intersections group both at home and abroad, this paperdeeply analyzes the diffusion mechanism of the intersections group and the analysis ofthe impact on traffic capacity. Using complex network to study the complexity of theurban road network, and the intersections group is put forward based on thecharacteristics of the complex network community. Basing upon the discontinuous traffic flow characteristics of the intersection in intersections group, applying the traffic wavetheory and the blockage effect on the upstream of road traffic flow with different queuelength, considering the practical operation of vehicles at the same time, analyzing theinfluence of queue length on the traffic capacity of the trunk road and law of intersection,providing the theoretical basis for the research of intersections group of the traffic stateidentification and signal control and guidance system of collaborative research.
In the thesis, the important role of the traffic state of intersection-group in trafficsignal control and dynamic traffic guidance is considered, takes the traffic statediscrimination of intersections group as the research object. First study on the base oftraffic state discrimination-traffic parameters prediction, analysis carefully of thecharacteristics of Elman neural network on the basis of analyzing the characteristics ofshort-time traffic flow, the method of section average velocity short-term forecast using adynamic recurrent neural network based on the variable gain factor is put forward, theeffectiveness of the method is verified using real-time data. On the base of obtainingaccurate future traffic parameters, the idea of the new basic unit of intersection-group isput forward, what is taking key intersection and related sections as a basic unit, usingintersection saturation, the section average speed and time occupancy for combinationdiscrimination of the basic unit traffic state, the model is verified based on the measureddata of the intersection of the Renmin Street and the Ziyou Road and its south importsections, results show that the combination discriminant method can improve theidentification precision, reduce the error brought by the single parameter. And the queuelength calculation formula of oversaturated intersections is given, it can provide accuratesection length for travel time prediction and convenient dynamic traffic informationcalculation in traffic guidance.
The cooperation optimized model for traffic signal control and traffic guidance ofintersections group is the focus of this paper, the relationship of traffic signal control andtraffic guidance as well as the impact on traffic flow are taken into account, thecollaborative feasibility of intersections group is analyzed, the driver route choicebehavior is studied from the driver's travel experience, the VMS traffic state guidanceinformation and the car terminal optimal path information based the prospect theory, andon this basis, the registration system optimal dynamic traffic assignment model is set up.Eventually set up the double layer cooperation model of traffic signal control dynamictraffic guidance of intersections group, the effectiveness of the model is verified usingthe simulative network.
On the base of collaborative model in chapter four, the information demand of thedrivers in the process of travel and the choice behavior of drivers for traffic guidanceinformation are taken into account in the fifth chapter. According to the characteristics ofthe intersections group, starting from the actual survey data, under the intersections group,research the publish content of traffic signal control and traffic guidance collaborativeinformation, information distribution way, collaborative information optimization ofrescue vehicles and information collaborative management plans in accidents and so onto realize the quick reach of rescue vehicles after emergency incidents, and eventuallyreach the purposes of emergency aid. Finally the collaborative management of buses andrescue vehicles under the emergency incidents is verified using the simulation method.
引文
[1]栗红强.城市交通控制信号配时参数优化方法研究[D].长春:吉林大学交通学院,2004.
[2] Turner I F, Shannon G F. Micro master-the co-ordination of small groups ofintersections[J]. Institute of Transportation Engineers,1978,19(4):171-173.
[3] Jee-Hyong Lee, Hyung Lee-Kwang. Distributed and cooperative fuzzy controllers fortraffic intersections group[J]. IEEE Transactions on Systems Man and Cyberneticspart C: Applications and Reviews,1999,29(2):263-271.
[4]皮钰鑫.信号控制交叉口群交通组织适应性研究[D].长沙:长沙理工大学交通运输工程学院,2009.
[5]杭明升.城市道路交叉口群实时自适应控制若干理论与方法研究[D]．上海:同济大学交通运输工程学院,2002.
[6]傅惠,徐建闽,卢凯.基于粒子群优化的关联交叉口群信号控制策略研究[J].交通与计算机,2007,25(3):23-26.
[7]任敏.饱和状态下交叉口群交通控制策略与配时优化研究[D].南京:东南大学交通学院,2010.
[8] Longley D. A control strategy for a congested computer-controlled traffic network[J].Transportation Research2.1968:391-408.
[9] May A. D., Montgomery F. O., Quinn D. J. Control of congestion in highly-congestednetworks[C], Proceedings of CODTU IV conference, Jakarta,1988
[10]吴洋.干道过饱和交叉口群的实时交通控制策略研究[D].成都:西南交通大学,2008.
[11]李岩,过秀成,杨洁等.过饱和状态交叉口群信号控制机理及实施框架[J].交通运输系统工程与信息,2011,11(4):28-34.
[12]高云峰,胡华,韩皓,杨晓光.城市道路交叉口群信号协调控制多目标优化与仿真[J].中国公路学报.2012,25(6):129-135.
[13]高云峰,胡华,杨晓光.交叉口群协调控制相位差优化模型研究[J].2006年第二届中国智能交通年会论文集,2006:329-333.
[14]傅惠,徐建闽,卢凯.基于粒子群优化的关联交叉口群信号控制策略研究[J].交通与计算机,2007,25(3):23-26.
[15]徐建闽,周沛,刘轼介.区域交叉口群协调控制方法研究[J].第七届中国智能交通年会优秀论文集-智能交通技术,2012:223-231.
[16]牟海波,俞建宁.城市交叉口群交通信号控制研究[J].兰州交通大学学报,2011,30(6):106-110.
[17]刘新杰.城市交通控制系统与交通流诱导系统协同策略研究[D].长春:吉林大学交通学院,2007.
[18]周八益,李琰,周溪召.动态交通控制-交通分配组合模型的求解算法研究[J],交通运输系统工程与信息,2003,3(1):55-61.
[19] Allsop R. E. Some possibilities for using traffic control to influence trip distributionand route choice[J]. Proceedings of the6th international symposium on traffic andtransportation theory,1974:345-375.
[20] Shimizu H, Kobayashi M, Yonezaw Y. A route guidance system of a trafficnetwork[C]. Proceedings of the34th SICE Annual Conference: International SessionPapers,1995:1549-1552.
[21]孙喜梅.城市路网实时动态交通信息的组合预测模型和方法研究[D].长春:吉林大学交通学院,2002.
[22] Gartner, Nathan H. Integration of dynamic traffic assignment with real-time trafficadaptive control system[J]. Transportation Research Record,1998,16(44):150-156.
[23] Cantarella, C. M. Shield, J. M. Anderson. Assignment in the Integration of UrbanTraffic Control and Dynamic Route Guidance in Urban Traffic Networks DynamicFlow Modeling and Control[J]. Springer-Verlag,1991:39-57.
[24] Yang H, Yagars. Traffic assignment and signal control in saturated road networks[J].Transportation Research Part A,1995,29(2):125-139.
[25]保丽霞.基于信息集成的城市交通流诱导与交通控制协同的关键理论及技术研究[D].长春:吉林大学交通学院,2006.
[26] Fisk, C. S. Optimal signal controls on congested networks[C].9th InternationalSymposium on Transportation and Traffic Theory,1984:197-216.
[27] Chen O. J., Ben-Akiva M. E. Game theoretic formulations of interaction betweendynamic traffic control and dynamic traffic assignment[J]. Transportation ResearchRecord,1998,1617:179-188.
[28]徐立群.城市交通流诱导与控制一体化理论和模型研究[D].长春:吉林大学交通学院,2000.
[29]陈昕.基于协同学的城市交通控制与诱导系统协同的理论与方法研究[D].长春:吉林大学交通学院,2006.
[30]王亮,马寿峰,李艳君.对交通控制与诱导结合研究的评述[J].系统工程,2004,22(2):56-62.
[31]王亮,马寿峰,贺国光.一种交通控制与诱导递阶协调优化模型[J].系统工程理论与实践,2004,(6):126-133.
[32]李瑞敏,史其信.基于多智能体系统的城市交通控制与诱导集成化研究[J].公路交通科技,2004,21(5):109-112.
[33]韩志新,魏连雨.交通管理中控制系统与诱导系统协调理论探索[J].河北工业大学学报,2003,32(1):26-30.
[34]方锦清.网络科学的理论模型探索及其进展[J].科技导报(北京),2006,24(12):67-72.
[35]胡一竑.基于复杂网络的交通网络复杂性研究[D],上海:复旦大学管理学院,2008.
[36] Amaral L A N, Scala A, Barthelemy M, Stanley H E. Classes of small-worldnetworks[J]. Proceedings National Academy of Sciences,2000,97:11149-11152.
[37] Latora V., Marchiori M. Is the Boston subway a small-world networks?[J]. PhysicaA,2002,314(l):109-113.
[38] Sen P., Dasgupta S., Chatterjee A., Sreeram P. A., Mukherjee G., Manna S. S.Small-world properties of the Indian railway network[J]. Physical Review E,2003,67:036106.
[39] Jiang B, Claramunt C. Topological analysis of urban street networks[J].Environment and Planning B,2004,31:151-162.
[40] Sienkiewicz J, Holyst J A. Public transport systems in Poland: from Bialystok toZielona Gora by bus and tram using universal statistics of complex networks[J]. ActaPhysica Poloniea B,2005,36(5):1771-1778.
[41] Chowell G, Hyman J M, Eubank S, Castillo-Chavez C. Sealing laws for themovement of People between locations in a large city[J]. Physical Review E,2003,68:066102.
[42] Montis A D, Barthelemy M, Chessa A, vespignanni A. The structure of inter-urbantraffic: A weighted network analysis[J]. Environment and Planning B: Planning andDesign,2007,34:905-924.
[43] Seaton K A, Hackett L M. Station, trains and small-world networks[J]. Physica A,2004,339:635-644.
[44] Wu J. J., Z. Y. Gao and H. J. Sun. Urban transit system as a scale-free network[J].Modern Physics Letters B,2004,18(19):1043-1049.
[45] Zhao Xiao-Mei, Gao Zi-You. Topological effect on the performance oftransportation networks[J]. Chinese Physical Letter,2007,24(1):283-286.
[46] Wu J. J., Gao Z. Y, and Sun H. J. Complexity and efficiency of Beijing transitnetwork[J]. International Journal of Modern physics B,2006,20(15):2129-2136.
[47]高自友,赵小梅,黄海军,毛保华.复杂网络理论与城市交通系统复杂性问题的相关研究[J].交通运输系统工程与信息,2006,6(3):41-47.
[48]吴建军.城市交通网络拓扑结构复杂性研究[D].北京:北京交通大学交通运输学院,2008.
[49]胡君辉,徐新平,杨永栩.三组城市公共汽车运输网的小世界性质[J].广西师范大学学报（自然科学版）,2006,24(2):10-14.
[50]何胜学，范炳全.从公交线网的生成机理看复杂网络的多样性[J].系统工程学报,2007,22(6):599-606.
[51]李英,周伟,郭世进.上海公共交通网络复杂性分析[J].系统工程学报,2007,25(l):38-41.
[52] Jordi Duch, Alex Arenas. Community detection in complex networks using externaloptimization[J]. Physical Review E,2005,72:027104.
[53] Girvan M, Newman M E J. Community structure in social and biologicalnetworks[J]. Proceedings of the National Academy of Sciences, U.S.A.,2002,99(12):7821-7826.
[54] H. Jeong, B. Tombor, R. Albert, Z. N. Oltvai, A.-L. Barabasi. The large-scaleorganization of metabolic networks[J]. Nature,2000,407:651-654.
[55]何毅.互联网中存在的群落结构及其对物联网发展的促进作用[J].信息与电脑,2011,1:74-75.
[56] GuimeràR, Danon L, Guilera A D, Giralt F, Arenas A. Self-similar communitystructure in a network of human interactions[J]. Physical Review E,2003,68:065103.
[57] GuimeràR, Mossa S, Turtschi A, Amaral L A N. The world-wide air transportationnetwork: anomalous centrality, community structure, and cities’ global roles[C].Proceedings of the National Academy of Sciences U.S.A.,2005,102(22):7794-7799.
[58] Porta S, Crucitti P, Latora V. The network analysis of urban streets: a dualapproach[J]. Physica A,2006,369(2):853-866.
[59]陆化普,朱军,王建伟.城市轨道交通规划的研究与实践[M].北京:中国水利水电出版社,2001.
[60] Li K P, Gao Z Y. Detecting and describing the modular structures of weightednetworks[J]. Chinese Physics,2007,16(8):2304-2309.
[61] Newman M. E. J. The structure and function of complex networks[J]. Society forIndustrial and applied Mathematics Review,2003,45(2):167-256.
[62] Clauset A, Newman M E J, Moore C. Finding community structure in very largenetworks[J]. Physical Review E,2004,70(6):066111.
[63]刘婷,胡宝清.基于聚类分析的复杂网络中的社团探测[J].复杂系统与复杂性科学,2007,4(l):28-35.
[64]崔迪.群落结构的复杂网络及其交通行为研究[D].北京:北京交通大学交通运输学院,2009.
[65] Newman M. E. J. Fast algorithm for detecting community structure in networks[J].Physical review E,2004,69(6):066133.
[66]刘婷,胡宝清.基于聚类分析的复杂网络中的社团探测[J].复杂系统与复杂性科学,2007,4(l):28-35.
[67] Pothen A, Simon H, Liou K P. Partitioning sparse matrices with eigenvectors ofgraphs[J]. SIAM Journal on Matrix Analysis and Applications,1990,11(3):430-452.
[68] Newman M. E. J., Girvan M. Finding and evaluating community structure innetworks[J]. Physical Review E,2004,69(2):026113.
[69]吴建军.城市交通网络拓扑结构复杂性研究[D].北京:北京交通大学交通运输学院,2008.
[70]汪小帆,陈关荣,李翔.复杂网络理论及其应用[M].北京:清华大学出版社,2006.
[71]宁宣熙.堵塞流理论及其应用[M].北京:科学出版社,2005.
[72]王进,史其信.短时交通流预测模型综述[J].智能交通,2005,1(6):92-98.
[73]冮龙辉.城市道路交通状态判别及拥堵扩散范围估计方法研究[D].长春:吉林大学交通学院,2007.
[74] C. J. Lan and S. P. Miaou. Real-time prediction of traffic flows using dynamicgeneralized linear models[J], Transportation Research Record,1999,1678:168-178.
[75] J. Kwon, B. Coifman and P. Bickel, Day-to-day travel time trends and travel timeprediction from loop detector data[C]. Transportation Research Record,2000,1717:120-129.
[76] Sun Hong-yu, Henry X. Liu, Heng Xiao, Rachel R. He and Bin Ran. Use of LocalLinear Linear Regression Model for Short-Term TrafficForecasting[J].Transportation Research Record,2002,1836:143-150.
[77] J. Rice, E.van Zwet. A simple and effective method for predicting travel times onfreeways[J]. IEEE Transaction on Intelligent Transportation Systems,2004,5(3):200-207.
[78] M. Sabry, H. Abd-El-Latif and N. Badra. Comparison between regression andARIMA models in forecasting traffic volume[J]. Australian journal of basic appliedsciences,2007,1(2):126-136.
[79] B. L. Smith, M. J. Demetsky. Traffic flow forecasting: Comparison of modelingapproaches[J]. Journal of transportation engineering-ASCE,1997,123(4):261-266.
[80] Chun-Hsin Wu, Jan-Ming Ho, Der-Tsai Lee. Travel-time prediction with supportvector regression[J], IEEE Transaction on Intelligent Transportation Systems,2004,5(4):276-281.
[81] B. M. Williams, P. K. Durvasula, D. E. Brown. Urban freeway traffic flowprediction: Application of seasonal autoregressive integrated moving average andexponential smoothing models[J], Transportation Research Record,1998,1644:132-141.
[82]冯金巧,杨兆升,张林,董升.一种自适应指数平滑动态预测模型[J].吉林大学学报（工学版）,2007,37(6):1284-1287.
[83]刘廷新,李振宇.指数平滑法在交通参数短期预测中的应用[J],山东交通学院学报,2002,10(3):20-22.
[84]姜桂艳.道路交通状态判别技术与应用[M].北京:人民交通出版社,2004.
[85] Hu Jian-ming, Song Jing-yan, Yu Guo-qiang, et al. A novel networked trafficparameter forecasting method based on Markov chain mode[C]. Proceedings of theIEEE International Conference on Systems Man and Cybernetics,2003:3595-3600.
[86]盖春英,裴玉龙.公路货运量灰色模型-马尔柯夫链预测方法研究[J].中国公路学报,2003,16(3):113-116.
[87]陈淑燕,王炜,瞿高峰.短时交通量时间序列的小波分析-模糊马尔柯夫预测方法[J],东南大学学报（自然科学版）,2005,35(4):637-640.
[88] P. Ross. Exponential filtering of traffic data[J]. Transportation Research Record.1982,869:43-49.
[89] I. Okutani, Y. J. Stephanedes. Dynamic prediction of traffic volume through Kalmanfiltering theory[J]. Transportation Research Part B: Method.1984,18(1):1-11.
[90] Ye Z., Zhang Y., Middleton D. Unscented Kalman filter method for speed estimationusing single loop detector data[J]. Transportation research record, Washington, D.C.,2006,1968:117-125.
[91]杨兆升,朱中.基于卡尔曼滤波理论的交通流量实时预测模型[J].中国公路学报,1999,12(3):63-67.
[92]李嘉,刘小兰.基于卡尔曼滤波理论的交通信息预测模型及其应用[J].中南公路工程,2005,30(4):108-109.
[93] Florio L., Mussone L. Neural network models for classification and forecasting offreeway traffic flow stability[J]. Control Engineering Practice,1996,4(2):153-164.
[94] H. R. Kirby, S. M. Watson, M. S. Dougherty. Should we use neural network orstatistical models for short-term motorway traffic forecasting[J], InternationalJournal of Forecasting,1997,13(1):43-50.
[95] B. Park, C. J. Messer and T. Urbanik. Short-term freeway traffic volume forecastingusing radial basis function neural network[J].Transportation Research Record,1998,1651:39-47.
[96] B. Park. Hybrid neural fuzzy application in short-term freeway traffic volumeforecasting[J]. Transportation Research Record,2002,1802:190-196.
[97] Chai Quek, Michel Pasquier, Bernard Boon Seng Lim. POP-TRAFFIC: A novelfuzzy neural approach to road traffic analysis and prediction[J]. IEEE Transactionson Intelligence Transportation Systems,2006,7(2):133-146.
[98]初连禹,杨兆升.基于神经网络的实时交通流量模糊预测[J],中国公路学报,1998, S1:78-82.
[99]杨兆升,谷远利.实时动态交通流预测模型研究[J].公路交通科技,1998,15(3):4-7.
[100]徐启华,丁兆奎,毕训银.一种基于动态递归神经网络的交通流量实时预测方法[J],淮海工学院学报(自然科学版),2003,12(4):14-17.
[101]于万霞,杜太行,郑宏兴,于越.基于粒子群的模糊神经网络交通流量预测[J].微计算机信,2008,24(21):232-233.
[102]杨兆升,朱中.基于BP神经网络的路径行程时间实时预测模型[J].系统工程理论与实践,1999,8:59-64.
[103]董春娇,邵春福.基于Elman神经网络的道路网短时交通流预测方法[J].交通运输系统工程与信息,2010,10(1):145-151.
[104] Abdalla M. I. Digital detection techniques via Elman neural network[J]. Journal ofEngineering and Applied Science,2002,49(6):1197-1208.
[105] X. Z. Gao, S. J. Ovaska. Genetic Algorithm Training of Elman Neural Network inMotor Fault Detection[J]. Neural Compute and Application,2002,11:37-44.
[106] Pham, D. T., Liu, X. Training of Elman networks and dynamic system modeling[J].International Journal of Systems Science,1996,27(2):221-226.
[107]杨庆芳,张彪,高鹏.基于改进动态递归神经网络的交通量短时预测方法[J].吉林大学学报（工学版）,2012,42(4):887-891.
[108]夏克文,李昌彪,沈钧毅.前向神经网络隐含层节点数的一种优化方法[J].计算机科学,2005,10(32):143-145.
[109]丛爽.面向MATLAB工具箱的神经网络理论与应用[M].北京:中国科学技术大学出版社,2003.
[110]张秀媛,达庆东,张国伍.公路自动事件检测技术[J].系统工程理论与实践,2001,21(6):118-124.
[111] Dudek C L, Messer C J, Nuckles N B. Incident Detection on Urban Freeways[C].Transportation Research Record, Washington D. C.: TRB,1974:12-24.
[112] Persaud B. N., Hall F. L. Catastrophe Theory and Patterns in30-second FreewayTraffic Data-Implication for incident detection[J]. Transportation Research Part A,1990,23(2):103-113.
[113]於毅.城市道路交通状态判别方法研究[D].北京:北京交通大学交通运输学院,2007.
[114]姜紫峰,刘小坤.基于神经网络的交通事件检测算法[J].西安公路交通大学学报,2000,20(3):67-69.
[115]郭恒明,张鹏飞.基于环形线圈的城市道路段交通异常自动检测方法研究[J].上海公路,2001,4:80-84.
[116]彭春露,彭国雄.城市道路交通异常事件管理系统设计[J].交通运输系统工程与信息,2002,2(3):19-23.
[117]杨兆升,杨庆芳,冯金巧.基于模糊综合推理的道路交通事件识别算法[J].公路交通科技,2003,20(4):92-94.
[118]肖永来.基于SCATS采集数据的城市道路交通状态判别技术研究[J].中国交通信息产业,2005,6:39-41.
[119]庄斌,杨晓光,李克平.道路交通拥挤事件判别准则与检测算法[J].中国公路学报,2006,19(3):82-86.
[120]贾森.基于实时信息的城市道路交通状态判别方法研究[D].北京:北京交通大学交通运输学院,2007.
[121]姜桂艳,郭海峰,吴超腾.基于感应线圈的城市道路交通状态判别方法[J],吉林大学学报（工学版）,2008,38(S1):37-42.
[122]张晓亮,张可,刘浩,李静,陈希.基于FCM-粗糙集的城市快速路交通状态判别[J].系统工程,2010,28(8):75-80.
[123]陆化普.解析城市交通[M].北京:中国水利水电出版社,2001.
[124]姜桂艳.道路交通状态判别技术与应用[M].北京:人民交通出版社,2004.
[125]张建华,林航飞.信号交叉口车辆两次停车率研究[J].交通科技,2004,3:96-97.
[126]哈肯.信息与自组织—复杂系统的宏观方法[M].成都:四川教育出版社,1988.
[127] Kerner B. S., Rehborn H. Experimental features and characteristics of traffic jam[J].Phys. Rev. E,1996,53:1297-1300.
[128] Kerner B. S., Konhauser, P. Experimental properties of phase transition in trafficflow[J]. Phy. Rev. Lett,1997,49:4030-4033.
[129]薛郁.交通流的建模、数值模拟及其临界相变行为的研究[D].上海:上海大学理学院,2002.
[130] Katsikopoulos K.V., Fisher D.L., Duse-Anthony Y，Duffy S.A. Risk attitudereversals in drivers’ route choice when range of travel time is provided[J]. HumanFactors,2002,44(3):466-473.
[131] Bogers E A I, Zuylen H J van. The importance of reliability in route choices infreight transport for various actors on various levels[C]. Proceeding EuropeanTransport Conference2004. France: Strasbourg,2004,16:136-154.
[132] Satoshi Fujii, Ryuichi Kitamura. Drivers’ mental representation of travel time anddeparture time choice in uncertain traffic network conditions[J].Networks andSpatial Economics,2004,4(3):243-256.
[133] Erel Avineri. A cumulative prospect theory approach to passengers behaviormodeling: waiting time paradox revisited[J]. Journal of Intelligent TransportationSystems,2004,8(4):195-204.
[134] Senbil M, Kitamura R. Reference points in commuter departure time choice: aprospect theoretic test of alternative decision frames[J]. Journal of IntelligentTransportation Systems,2004,8:19-31.
[135]赵凛,张星臣.基于前景理论的路径选择行为建模及实例分析[J].土木工程学报,2006,6(2):42-46.
[136]徐红利,周晶,陈星光.基于前景理论的路径选择行为规则分析与实证[J].交通运输系统工程与信息,2007,(6):95-101.
[137]杨志勇.基于前景理论的出发时刻和出行路径选择模型研究[D].哈尔滨:哈尔滨工业大学交通科学与工程学院,2007.
[138]王燕.基于期望效用理论与前景理论的出行决策模型对比研究[D].成都:西南交通大学交通运输与物流学院,2011.
[139] Kahneman D, Tversky A. Prospect theory: Ananalysis of decisions under risk[J].Econometric,1979,47:313-327.
[140]薛求知,黄佩燕等.行为经济学-理论与应用[M].上海:复旦大学出版社,2003.
[141] Tversky A., Kahneman D. Advances in prospect theory: cumulative representationof uncertainty[J]. Journal of Risk and Uncertainty,1992,9:195-230.
[142]陆化普,史其信,殷亚峰.动态交通分配理论的回顾与展望[J].公路交通科技,1996,13(2),34-43.
[143]马寿峰,卜军峰,张安训.交通诱导中系统最优与用户最优的博弈协调[J].系统工程学报,2005,20(1)：30-37.
[144] Wie B W, Tobin R L. Comparison of system optimum and user equilibriumdynamic traffic assignments with schedule delays[J]. Transportation Research Part C,1995,36:389-411.
[145] Adler J L, Blue V J. Toward the design of intelligent traveler informationsystems[J]. Transportation Research Part C,1998,3(6):157-172.
[146]尹涛,梁谦.面向ATMS的交通信息系统研究[J].公路,2009,3(3):87-91.
[147] Linda Ng Boyle, Fred Mannering. Impact of traveler advisory systems on drivingspeed: some new evidence[J]. Transportation Research Part C,2004,12(1):57-72.
[148]周元峰.基于信息的驾驶员路径选择行为及动态诱导模型研究[D].北京：北京交通大学交通运输学院,2007.
[149] Piyushimita Thakuriah, Ashish Sen. Quality of information given by advancedtraveler information systems[J]. Transportation Research Part C,1996,4(5):249-266.
[150] Hato E, Taniguchib M, Sugiec Y, et al. Incorporating an information acquisitionprocess into a route choice model with multiple information sources[J].Transportation Research Part C,1999,5(7):109-129.
[151]孙凤霞.城市区域交通可变信息板布点方法研究[D].长春:吉林大学交通学院,2008.
[152]张春勤.道路交通动态信息发布策略研究[D].长春:吉林大学交通学院,2011.