大型公共场所行人交通状态评价及其应急疏散方法研究
|
摘要
随着社会经济的持续快速发展，我国城市现代化、智能化水平明显有所提升，当前城市先进的道路交通设施使得人们的交通出行特征、交通出行条件发生显著改变，比如人们的交通出行次数、交通出行时长都有所增长，同时我国很多城市人口密度较大，居民快速聚集的几率大大提高。
对于道路交通管理方面的研究，以往给予车辆交通问题给予更多重视，行人交通问题往往却被忽略。近几年来，随着我国各地大型公共设施（如步行街、体育场、电影院、展览馆等）纷纷落成以及各种大型活动（如演唱会、足球赛、签售会等）举办，行人交通监控、管理、疏散等方面的研究开始成为热点领域。在人们日常生活中，交通高峰时段往往容易出现严重的车辆拥堵现象。与此相似，在购物中心、人行横道、地下通道、火车站、地铁站、步行街等地点，经常看到行人交通拥堵，行人交通拥堵也会造成人们出行延误，而且当行人密度过大时，行人交通拥堵容易形成重大安全事故。
本文充分借鉴智能交通系统领域已有的相关理论与方法，结合“西单地区人员密集场所预警系统及示范工程”的环境特点以及行人交通参数数据的规律和特点，设计了一类大型公共场所的行人交通状态评价方法以及行人交通应急疏散模型。主要开展了以下几个方面的研究工作：
1）行人交通特性分析及状态分类。根据行人交通数据时间序列稳定性的差异，定义了数据序列的三个特性，即长期趋势性、短期现势性及随机波动性；为了能给行人交通状态的判断提供更具可靠性的基础信息，采用不同星期同一天数据设计了行人交通数据长期发展模式的构建方法；为了便于根据不同交通状况制定更有针对性的行人交通管理措施，通过分析行人交通数据时间序列的特征进行了行人交通状态分类。
2）行人交通数据预处理。为了提高行人交通状态在线评价系统的输入数据质量水平，结合行人交通数据时间序列变化规律及其特征，依次设计了行人交通丢失数据的处理方法、错误数据的处理方法、时间序列的滤波方法以及不同时间尺度数据的合成方法。
3）行人交通事件自动检测算法研究。针对行人交通数据的独特性，首先从预测方法和标准偏差值过小引发误警两个方面对标准偏差法进行了改进，然后设计了一种基于支持向量机和标准偏差法的交通事件自动检测组合算法，最后设计了行人交通事件的报警机制。
4）行人交通拥堵检测与预测方法研究。首先研究了行人交通拥堵自动报警方法，包括行人交通拥堵自动检测算法以及行人交通拥堵自动报警机制，重点研究了行人交通数据多步预测方法，共分为三个方案，即长期趋势多步预测、短期现势多步预测以及二者的组合预测方法，并且设计了行人交通拥堵状态的预警机制。
5）行人交通应急疏散方法路径预案研究。行人交通事件或行人交通拥堵的产生非常容易形成人群的聚集，为了避免形成严重的行人安全事故或者减小造成人员伤亡以及财产损失，通过构建行人安全疏散原则和分析行人疏散影响因素，充分考虑广场、步道、楼梯等各种大型公共场所行人交通应急疏散的特点，分别建立了行人交通应急疏散模型以及大型公共场所整体结构应急疏散模型。
The continuous economic growth and the accelerating urbanization in Chinabrought about a transparent change of city transportation system. Travel frequencyand travel time increased. Most cities in the country witnessed a growing populationand density.
Previous studies had dealt considerably on issues of automobile transportationwhile neglected the problems created by pedestrians. However, the number ofpedestrians was not fewer just because of more automobiles on road. This years, largepublic venues such as arenas, cinemas and assembly centers were emerging andbecame concentrations of participants for concert, football match and book signingand others. The monitoring, management and evacuation of this population hasbecome an academic hotspot. Similar to rush hour congestions of vehicles,pedestrians were also worth of critical consideration for safety in public spaces suchas shopping malls, zebra crossings, subways, railway and light rail stations.
This paper is a redesign of on-line assessment which is aimed at meeting certainrequirements for the pedestrian flow at the sites of specific public areas. Previousstudies of intelligent transport system and thoughts and theories were employed. Validparameter data of a densely populated area of Beijing was examined as a model forpedestrian evacuation and danger preventives. The paper engaged following aspects.
1) Parameters pertinent to pedestrian flow, density and regional distributionwere collected to probe pedestrian traffic features. After that, in reference withstability variance of pedestrian time series, three typical features of pedestrian flowwere defined, i.e. the long-term tendency, short-term trend and randomized volatility.To suit specific requirements variable to the environment, categorization of the statusquo is necessarily to make in accordance with different presumptions. For sustainableprogress, the modes of normality and travelling were designated. As far as transportefficiency is concerned, two modes of smoothness and congestion were also set up.
2) Pedestrian traffic data preprocessing. In order to improve the state ofpedestrian traffic input data quality level of the online evaluation system, combiningthe pedestrian traffic time series data change rules and characteristics, design the lostdata in order to identify and repair method, error data identification and repair method,time series smoothing method and large time scale synthesis method.
3) In use of support vector machine, or SVM, an automatic calculation ofpedestrian incidents was designed. After that, an abrupt change index mechanism wasconstructed and a classification of pedestrian activity was sought after. Therefore analarm mechanism was designed including an incessant checking system and onebased on probability estimate.
4) Pedestrian traffic congestion state short-term forecast method research. Inorder to provide reliable data for the pedestrian traffic congestion state warning,mainly studies the pedestrian traffic data multi-step prediction method, which isdivided into three schemes, namely the multi-step prediction, long-term trend,short-term current multi-step prediction as well as the two combined forecastingmethod. Then, design the pedestrian traffic congestion state of early warningmechanism.
5) Pedestrian traffic emergency evacuation method research. Pedestrian trafficevents or pedestrian traffic is very easy to form a crowd gathered, at this point, thecrowd there exist great risk factors. If let developments, or if other events occur, suchas a pedestrian collisions, pushing, shoving down, it's easy to form serious pedestriansafety accident, resulting in casualties and property losses. Therefore, scientificpedestrian traffic emergency evacuation mechanism can guide the pedestrian quicklyleave the site, or avoid blindly into the area, traffic events or pedestrian trafficinfluence quickly narrowed down, eventually to be able to properly solve. By buildinga pedestrian safety evacuation standard, the factors influencing the pedestrianevacuation, and fully consider the characteristics of pedestrian traffic emergencyevacuation of all kinds of large-scale public places, the pedestrian traffic emergencyevacuation model was established, on the basis of the traffic incident or pedestriantraffic congestion situation, to ensure the safety of pedestrian traffic.
对于道路交通管理方面的研究，以往给予车辆交通问题给予更多重视，行人交通问题往往却被忽略。近几年来，随着我国各地大型公共设施（如步行街、体育场、电影院、展览馆等）纷纷落成以及各种大型活动（如演唱会、足球赛、签售会等）举办，行人交通监控、管理、疏散等方面的研究开始成为热点领域。在人们日常生活中，交通高峰时段往往容易出现严重的车辆拥堵现象。与此相似，在购物中心、人行横道、地下通道、火车站、地铁站、步行街等地点，经常看到行人交通拥堵，行人交通拥堵也会造成人们出行延误，而且当行人密度过大时，行人交通拥堵容易形成重大安全事故。
本文充分借鉴智能交通系统领域已有的相关理论与方法，结合“西单地区人员密集场所预警系统及示范工程”的环境特点以及行人交通参数数据的规律和特点，设计了一类大型公共场所的行人交通状态评价方法以及行人交通应急疏散模型。主要开展了以下几个方面的研究工作：
1）行人交通特性分析及状态分类。根据行人交通数据时间序列稳定性的差异，定义了数据序列的三个特性，即长期趋势性、短期现势性及随机波动性；为了能给行人交通状态的判断提供更具可靠性的基础信息，采用不同星期同一天数据设计了行人交通数据长期发展模式的构建方法；为了便于根据不同交通状况制定更有针对性的行人交通管理措施，通过分析行人交通数据时间序列的特征进行了行人交通状态分类。
2）行人交通数据预处理。为了提高行人交通状态在线评价系统的输入数据质量水平，结合行人交通数据时间序列变化规律及其特征，依次设计了行人交通丢失数据的处理方法、错误数据的处理方法、时间序列的滤波方法以及不同时间尺度数据的合成方法。
3）行人交通事件自动检测算法研究。针对行人交通数据的独特性，首先从预测方法和标准偏差值过小引发误警两个方面对标准偏差法进行了改进，然后设计了一种基于支持向量机和标准偏差法的交通事件自动检测组合算法，最后设计了行人交通事件的报警机制。
4）行人交通拥堵检测与预测方法研究。首先研究了行人交通拥堵自动报警方法，包括行人交通拥堵自动检测算法以及行人交通拥堵自动报警机制，重点研究了行人交通数据多步预测方法，共分为三个方案，即长期趋势多步预测、短期现势多步预测以及二者的组合预测方法，并且设计了行人交通拥堵状态的预警机制。
5）行人交通应急疏散方法路径预案研究。行人交通事件或行人交通拥堵的产生非常容易形成人群的聚集，为了避免形成严重的行人安全事故或者减小造成人员伤亡以及财产损失，通过构建行人安全疏散原则和分析行人疏散影响因素，充分考虑广场、步道、楼梯等各种大型公共场所行人交通应急疏散的特点，分别建立了行人交通应急疏散模型以及大型公共场所整体结构应急疏散模型。
The continuous economic growth and the accelerating urbanization in Chinabrought about a transparent change of city transportation system. Travel frequencyand travel time increased. Most cities in the country witnessed a growing populationand density.
Previous studies had dealt considerably on issues of automobile transportationwhile neglected the problems created by pedestrians. However, the number ofpedestrians was not fewer just because of more automobiles on road. This years, largepublic venues such as arenas, cinemas and assembly centers were emerging andbecame concentrations of participants for concert, football match and book signingand others. The monitoring, management and evacuation of this population hasbecome an academic hotspot. Similar to rush hour congestions of vehicles,pedestrians were also worth of critical consideration for safety in public spaces suchas shopping malls, zebra crossings, subways, railway and light rail stations.
This paper is a redesign of on-line assessment which is aimed at meeting certainrequirements for the pedestrian flow at the sites of specific public areas. Previousstudies of intelligent transport system and thoughts and theories were employed. Validparameter data of a densely populated area of Beijing was examined as a model forpedestrian evacuation and danger preventives. The paper engaged following aspects.
1) Parameters pertinent to pedestrian flow, density and regional distributionwere collected to probe pedestrian traffic features. After that, in reference withstability variance of pedestrian time series, three typical features of pedestrian flowwere defined, i.e. the long-term tendency, short-term trend and randomized volatility.To suit specific requirements variable to the environment, categorization of the statusquo is necessarily to make in accordance with different presumptions. For sustainableprogress, the modes of normality and travelling were designated. As far as transportefficiency is concerned, two modes of smoothness and congestion were also set up.
2) Pedestrian traffic data preprocessing. In order to improve the state ofpedestrian traffic input data quality level of the online evaluation system, combiningthe pedestrian traffic time series data change rules and characteristics, design the lostdata in order to identify and repair method, error data identification and repair method,time series smoothing method and large time scale synthesis method.
3) In use of support vector machine, or SVM, an automatic calculation ofpedestrian incidents was designed. After that, an abrupt change index mechanism wasconstructed and a classification of pedestrian activity was sought after. Therefore analarm mechanism was designed including an incessant checking system and onebased on probability estimate.
4) Pedestrian traffic congestion state short-term forecast method research. Inorder to provide reliable data for the pedestrian traffic congestion state warning,mainly studies the pedestrian traffic data multi-step prediction method, which isdivided into three schemes, namely the multi-step prediction, long-term trend,short-term current multi-step prediction as well as the two combined forecastingmethod. Then, design the pedestrian traffic congestion state of early warningmechanism.
5) Pedestrian traffic emergency evacuation method research. Pedestrian trafficevents or pedestrian traffic is very easy to form a crowd gathered, at this point, thecrowd there exist great risk factors. If let developments, or if other events occur, suchas a pedestrian collisions, pushing, shoving down, it's easy to form serious pedestriansafety accident, resulting in casualties and property losses. Therefore, scientificpedestrian traffic emergency evacuation mechanism can guide the pedestrian quicklyleave the site, or avoid blindly into the area, traffic events or pedestrian trafficinfluence quickly narrowed down, eventually to be able to properly solve. By buildinga pedestrian safety evacuation standard, the factors influencing the pedestrianevacuation, and fully consider the characteristics of pedestrian traffic emergencyevacuation of all kinds of large-scale public places, the pedestrian traffic emergencyevacuation model was established, on the basis of the traffic incident or pedestriantraffic congestion situation, to ensure the safety of pedestrian traffic.
引文
[1]胡清梅.大型公共建筑环境中人群拥挤机理及群集行为特性的研究[D].北京交通大学，2006.
[2] May A D. Traffic Flow Fundamental [M].1st ed. New Jersey: Prentice Hall Inc,1990.
[3] Fruin J J. Pedestrian Planning and Design [M].2nd ed. Alabama: Elevator WorldInc,1987.
[4] Helbing D, Molnar P. Self-Organization Phenomena in Pedestrian Crowds [A] In:F. Schweitzer. Self-Organization of Complex Structures: From Individual toCollective Dynamics [M]. London: Gordon and Breach，1998.
[5] Helbing D. A Mathematical Model for the Behavior of Pedestrians [J]. BehavioralScience,1991,36:298-310.
[6] Blue V J, Adler J L. Emergent Fundamental Pedestrian Flows from CellularAutomata Micro-simulation [J]. Transportation Research Record,1998,1644:29-36.
[7] Hoogendoorn S P, Bovy P H L. Gas-kinetic modeling and simulation of Pedestrianflows [J]. Transportation Research Record,2000,1710:28-36.
[8] Klupfel H, Meyer-Konig T, Wahle J, Schreckenberg M. Microscopic simulation ofevacuation processes on passenger ships [M]. London: Springer,2000:198-224.
[9] Muramatsu M, Irie T, Nagatani T. Jamming transition in pedestrian counter flow[J]. Physical A,1999,267:487-498.
[10] Lam W H K, John F M, Herbert H. Pedestrian Flow Characteristics in HongKong [J].Transportation Research Record,1995,1487:56-62.
[11]史建港.大型活动行人交通特性研究[D].北京工业大学，2007.
[12]岳昊.基于元胞自动机的行人流仿真模型研究[D].北京交通大学，2008.
[13]马剑.相向行人流自组织行为机理研究[D].中国科学技术大学，2010.
[14]裴玉龙，马骥.实时交通数据的筛选与恢复研究[J].土木工程学报，2003（7）：78-83.
[15]姜桂艳，刚龙晖，张晓东等.动态交通数据故障识别与修复方法[J].交通运输工程学报，2004(1)：121-125.
[16]耿彦斌，于雷，赵慧. ITS数据质量控制技术及应用研究[J].中国安全科学学报，2005(1)：82-87.
[17]陈淑燕，王炜，李文勇.实时交通数据的噪声识别和消噪方法[J].东南大学学报(自然科学版)，2006(2)：322-325.
[18]秦玲，郭艳梅，吴鹏等.断面交通检测数据检验及预处理关键技术研究[J].公路交通科技(应用技术版)，2006(11)：39-42.
[19] Payne H J, Tignor S C. Freeway Incident-detection Algorithms based on DecisionTrees with States [C]. Transportation Research Board, Washington D.C.,1978, n.682, p.30-37.
[20] Dudek C L, Messer C J, Nuckles N B. Incident Detection on Urban Freeways [C].Transportation Research Board, Washington D.C.,1974, n.495, p.12-24.
[21] Cook A R, Cleveland D E. Detection of Freeway Capacity-Reducing Incidents byTraffic-Stream Measurements [J]. TRR,1974, n.495, p.1-11.
[22] Ahmed M. S. and A. R. Cook. Analysis of Freeway Traffic Time-Series DataUsing Box-Jenkins Techniques [J]. Transportation Research Record, No.722,National Research Council,1977:1-9.
[23] Levin M, Krause G M. Incident Detection: A Bayesian Approach [J].Transportation Research Record,1978, n.682, p.52-58.
[24] Persaud B N, Hall F L. Catastrophe Theory and Patterns in30-second FreewayTraffic Data-Implication for incident detection [J]. Transportation Research PartA,1990, v.23, n.2, p.103-113.
[25] Balke K., C. L. Dudek and C. E. Mountain. Using Probe-measured Travel Timeto Detect Major Freeway Incidents in Houston, Texas [J]. TransportationResearch Record, No.1554, TRB, National Research Council,1996:213-220.
[26]李文江.高速公路监控系统与基于小波分析的事件检测算法研究[D].西安：西安公路交通大学，1999.
[27]姜桂艳，温慧敏，杨兆升.高速公路交通事件自动检测系统与算法设计[J].交通运输工程学报，2001，1(1)：77-81.
[28]周雪铭.基于小波分析和神经网络的交通事件自动检测[J].公路与汽车，2002,(4)：17-19.
[29]李海峰，王炜，姜桂艳.基于SNN软件系统的神经网络交通事件自动检测算法[J].公路交通科技，2004，21(11)：89-92.
[30]李红伟.快速路动态交通数据质量评价与控制方法研究[D].长春：吉林大学，2009．
[31] Yasdi R. Prediction of Road Traffic using a Neural Network Approach[R]．Neural Compute&Applica1999:135-142．
[32]孙燕，陈森发，周振国.灰色系统理论在无检测器交叉口交通流量预测中的应用[J].东南大学学报（自然科学版），2002，32(2)：151-153．
[33] Brian L. Smith, Billy M. Williams, R. Keith Oswald. Comparison of parametricand nonparametric models for traffic flow forecasting [J]. TransportationResearch Part C,2002,(10):303-321．
[34] Yiannis Kamarianakis, Poulicos Prastacos. Space–time modeling of traffic flow[J]. Computers&Geosciences,2005,(31):119-133．
[35]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．
[36]翁小雄，杜高丽，谭国贤.基于时间占有率的短时交通预测模[J].控制工程，2005：102-104．
[37] Loukas D, Theodore T, Antony S. Adaptive hybrid fuzzy rule-based systemapproach for modeling and predicting urban traffic flow [J]. TransportationResearch Part C,2007:1-20．
[38] Xus J，Shi Z．Short-Time Traffic Flow Prediction Based on Chaos Time SeriesTheory [J]. Journal of Transportation Systems Engineering and InformationTechnology,2008,8(5):68-72.
[39]曹凤萍.城市道路交通参数短时多步预测方法研究[D].长春：吉林大学，2007．
[40]李明涛.快速路短时间尺度地点交通参数多步预测方法研究[D].长春：吉林大学，2009.
[41] Smith B L, Conklin J H. The use of local lane distribution patterns for theestimation of missing data in transportation management systems [C].Transportation Research Board, Washington DC,2002
[42] Zhong M, Lingras P, Sharma S. Estimation of missing traffic counts using factor,genetic, neural, and regression techniques[J]. Transportation Research,2004,12(2):139-166.
[43]徐健锐，李星毅，施化吉.处理缺失数据的短时交通流预测模型[J].计算机应用，2010，30（4）：1117-1120．
[44]王裙，吴兵，王靖阳. ITS实时检测交通数据的质量控制技术探讨[J].交通与运输，2005，（7）：30-32．
[45]张旭.面向交通运行状态评价的多源异质交通流数据融合技术方法研究[D].北京：北京交通大学，2008.
[46]邱奉翠.基于城市快速路和主干道多源交通检测信息的数据融合技术研究[D].北京：北京交通大学，2012.
[47]王敏会.相依随机变量的移动平均过程的完全收敛性[D].长春：吉林大学，2004.
[48]苏海忠.相依样本下移动平均过程的矩完全收敛[D].南宁：广西师范学院，2012.
[49]李颖.时间序列指数平滑算法的改进研究[D].阜新：辽宁工程技术大学，2009.
[50]张丽.基于动态指数平滑模型的粮食价格预测方法研究[D].郑州：河南工业大学，2011.
[51]曹凤萍.城市道路交通参数短时多步预测方法研究[D].长春：吉林大学，2008.
[52]王秋兰.短时交通参数多步预测方法研究[D].长春：吉林大学，2012.
[53]李琦.基于多源数据的交通状态监测与预测方法研究[D].长春：吉林大学，2013.
[54] Payne H J, Tignor S C. Freeway Incident Detection Algorithms Based onDecision Trees with States[J]. Transportation Research Record682, TRB,National Research Council, Washington D.C.,1978:30-37.
[55] Martin P, Perrin H, Hansen B. Incident detection algorithm evaluation [R].Technical Report UTL-0700-31, Utah Traffic Laboratory, July,2000.
[56]史新宏，蔡伯根．高速公路自动事件检测算法[J]．交通运输系统工程与信息，2011，37（8）：306-310．
[57] Dudek C L, Messer G M. Incident Detection on Urban Freeways[J].Transportation Research Record459, TRB, National Research Council,Washington D.C.,1974,(495):12-24.
[58] Cook A R, Cleveland D E. Detection of Freeway Capacdty Reducing Incidents byTraffic Stream Measurement[J]. Transportation Research Record, TRB,1974,(495):1-11.
[59] Martin P, Perrin H, Hansen B.Incident detection algorithm evaluation [R].Technical Report UTL-0700-31, Utah Traffic Laboratory, July,2000.
[60] Collins J F, Hopkins C M, Martin J A. Automatic Incident Detection. TRRLAlgorithms HIOCC and PATREG[R]. Supplementary Report526, Transport andRoad Research Laboratory, Crowthorne, England,1979.
[61] Ahmed S A, Cook A R. Application of Time-Series Analysis Techniques toFreeway Incident Detection[J]. Transportation Research Record682, TRB,National Research Council, Washington D.C.,1980,(841):19-21.
[62] Persaud B N, Hall F L, Hall L M. Congestion Identification Aspects of theMcMaster Incident Detection Algorithm[J]. Transportation Research Record,1990,(1287):167-175.
[63] Hsiao C H; Lin C T; Cassidy M. Application of Fuzzy Logic and NeuralNetworks to Automatically Detect Freeway Traffic Incidents[J]. Journal ofTransportation Engineering,1994,120(5):753-771.
[64] Ritchie S G; Cheu R L. Simulation of Freeway Incident Detection UsingArtificial Neural Networks[J]. Trasportation Research Part C,1993,1(3):313-331.
[65] Teng H, Qi Y. Application of wavelet technique to freeway incident detection[J].Transportation Research Part C: EmergingTechnologies,2003,11(3):289-308.
[66] Yuan F, Cheu R L. Incident detection using support vector machines[J].Transportation Research Part C: Emerging Technologies,2003,11(3):309-328.
[67]刘彬生.基于小波和傅立叶变换的道路交通量预测研究[D].哈尔滨：哈尔滨工业大学，2006.
[68]倪同和.道路交通规划关键指标预测方法研究[D].长春：吉林大学，2011.
[69]纪冬梅.基于组合模型的短时交通流预测方法[D].上海：华东理工大学，2012.
[70]喻丹.基于小波理论的短时交通流预测方法研究[D].长沙：长沙理工大学，2008.
[71]曹征.基于小波变换的交通流短时预测模型研究[D].北京：北京交通大学，2010.
[72]谭政.城市道路交通流预测及应用[D].广州：华南理工大学，2012.
[73]李存军.基于集成神经网络的城市道路交通流量融合预测研究[D].成都：西南交通大学，2004.
[74]李贞珍.神经网络在城市交通流预测模型中应用的研究[D].上海：上海交通大学，2007.
[75]王沥.基于改进神经网络的短时交通流预测研究[D].成都：电子科技大学，2012.
[76] Pauls J. Calculating evacuation times for tall buildings[J]. Fire Safety Journal,1987,12(3):213-236.
[2] May A D. Traffic Flow Fundamental [M].1st ed. New Jersey: Prentice Hall Inc,1990.
[3] Fruin J J. Pedestrian Planning and Design [M].2nd ed. Alabama: Elevator WorldInc,1987.
[4] Helbing D, Molnar P. Self-Organization Phenomena in Pedestrian Crowds [A] In:F. Schweitzer. Self-Organization of Complex Structures: From Individual toCollective Dynamics [M]. London: Gordon and Breach，1998.
[5] Helbing D. A Mathematical Model for the Behavior of Pedestrians [J]. BehavioralScience,1991,36:298-310.
[6] Blue V J, Adler J L. Emergent Fundamental Pedestrian Flows from CellularAutomata Micro-simulation [J]. Transportation Research Record,1998,1644:29-36.
[7] Hoogendoorn S P, Bovy P H L. Gas-kinetic modeling and simulation of Pedestrianflows [J]. Transportation Research Record,2000,1710:28-36.
[8] Klupfel H, Meyer-Konig T, Wahle J, Schreckenberg M. Microscopic simulation ofevacuation processes on passenger ships [M]. London: Springer,2000:198-224.
[9] Muramatsu M, Irie T, Nagatani T. Jamming transition in pedestrian counter flow[J]. Physical A,1999,267:487-498.
[10] Lam W H K, John F M, Herbert H. Pedestrian Flow Characteristics in HongKong [J].Transportation Research Record,1995,1487:56-62.
[11]史建港.大型活动行人交通特性研究[D].北京工业大学，2007.
[12]岳昊.基于元胞自动机的行人流仿真模型研究[D].北京交通大学，2008.
[13]马剑.相向行人流自组织行为机理研究[D].中国科学技术大学，2010.
[14]裴玉龙，马骥.实时交通数据的筛选与恢复研究[J].土木工程学报，2003（7）：78-83.
[15]姜桂艳，刚龙晖，张晓东等.动态交通数据故障识别与修复方法[J].交通运输工程学报，2004(1)：121-125.
[16]耿彦斌，于雷，赵慧. ITS数据质量控制技术及应用研究[J].中国安全科学学报，2005(1)：82-87.
[17]陈淑燕，王炜，李文勇.实时交通数据的噪声识别和消噪方法[J].东南大学学报(自然科学版)，2006(2)：322-325.
[18]秦玲，郭艳梅，吴鹏等.断面交通检测数据检验及预处理关键技术研究[J].公路交通科技(应用技术版)，2006(11)：39-42.
[19] Payne H J, Tignor S C. Freeway Incident-detection Algorithms based on DecisionTrees with States [C]. Transportation Research Board, Washington D.C.,1978, n.682, p.30-37.
[20] Dudek C L, Messer C J, Nuckles N B. Incident Detection on Urban Freeways [C].Transportation Research Board, Washington D.C.,1974, n.495, p.12-24.
[21] Cook A R, Cleveland D E. Detection of Freeway Capacity-Reducing Incidents byTraffic-Stream Measurements [J]. TRR,1974, n.495, p.1-11.
[22] Ahmed M. S. and A. R. Cook. Analysis of Freeway Traffic Time-Series DataUsing Box-Jenkins Techniques [J]. Transportation Research Record, No.722,National Research Council,1977:1-9.
[23] Levin M, Krause G M. Incident Detection: A Bayesian Approach [J].Transportation Research Record,1978, n.682, p.52-58.
[24] Persaud B N, Hall F L. Catastrophe Theory and Patterns in30-second FreewayTraffic Data-Implication for incident detection [J]. Transportation Research PartA,1990, v.23, n.2, p.103-113.
[25] Balke K., C. L. Dudek and C. E. Mountain. Using Probe-measured Travel Timeto Detect Major Freeway Incidents in Houston, Texas [J]. TransportationResearch Record, No.1554, TRB, National Research Council,1996:213-220.
[26]李文江.高速公路监控系统与基于小波分析的事件检测算法研究[D].西安：西安公路交通大学，1999.
[27]姜桂艳，温慧敏，杨兆升.高速公路交通事件自动检测系统与算法设计[J].交通运输工程学报，2001，1(1)：77-81.
[28]周雪铭.基于小波分析和神经网络的交通事件自动检测[J].公路与汽车，2002,(4)：17-19.
[29]李海峰，王炜，姜桂艳.基于SNN软件系统的神经网络交通事件自动检测算法[J].公路交通科技，2004，21(11)：89-92.
[30]李红伟.快速路动态交通数据质量评价与控制方法研究[D].长春：吉林大学，2009．
[31] Yasdi R. Prediction of Road Traffic using a Neural Network Approach[R]．Neural Compute&Applica1999:135-142．
[32]孙燕，陈森发，周振国.灰色系统理论在无检测器交叉口交通流量预测中的应用[J].东南大学学报（自然科学版），2002，32(2)：151-153．
[33] Brian L. Smith, Billy M. Williams, R. Keith Oswald. Comparison of parametricand nonparametric models for traffic flow forecasting [J]. TransportationResearch Part C,2002,(10):303-321．
[34] Yiannis Kamarianakis, Poulicos Prastacos. Space–time modeling of traffic flow[J]. Computers&Geosciences,2005,(31):119-133．
[35]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．
[36]翁小雄，杜高丽，谭国贤.基于时间占有率的短时交通预测模[J].控制工程，2005：102-104．
[37] Loukas D, Theodore T, Antony S. Adaptive hybrid fuzzy rule-based systemapproach for modeling and predicting urban traffic flow [J]. TransportationResearch Part C,2007:1-20．
[38] Xus J，Shi Z．Short-Time Traffic Flow Prediction Based on Chaos Time SeriesTheory [J]. Journal of Transportation Systems Engineering and InformationTechnology,2008,8(5):68-72.
[39]曹凤萍.城市道路交通参数短时多步预测方法研究[D].长春：吉林大学，2007．
[40]李明涛.快速路短时间尺度地点交通参数多步预测方法研究[D].长春：吉林大学，2009.
[41] Smith B L, Conklin J H. The use of local lane distribution patterns for theestimation of missing data in transportation management systems [C].Transportation Research Board, Washington DC,2002
[42] Zhong M, Lingras P, Sharma S. Estimation of missing traffic counts using factor,genetic, neural, and regression techniques[J]. Transportation Research,2004,12(2):139-166.
[43]徐健锐，李星毅，施化吉.处理缺失数据的短时交通流预测模型[J].计算机应用，2010，30（4）：1117-1120．
[44]王裙，吴兵，王靖阳. ITS实时检测交通数据的质量控制技术探讨[J].交通与运输，2005，（7）：30-32．
[45]张旭.面向交通运行状态评价的多源异质交通流数据融合技术方法研究[D].北京：北京交通大学，2008.
[46]邱奉翠.基于城市快速路和主干道多源交通检测信息的数据融合技术研究[D].北京：北京交通大学，2012.
[47]王敏会.相依随机变量的移动平均过程的完全收敛性[D].长春：吉林大学，2004.
[48]苏海忠.相依样本下移动平均过程的矩完全收敛[D].南宁：广西师范学院，2012.
[49]李颖.时间序列指数平滑算法的改进研究[D].阜新：辽宁工程技术大学，2009.
[50]张丽.基于动态指数平滑模型的粮食价格预测方法研究[D].郑州：河南工业大学，2011.
[51]曹凤萍.城市道路交通参数短时多步预测方法研究[D].长春：吉林大学，2008.
[52]王秋兰.短时交通参数多步预测方法研究[D].长春：吉林大学，2012.
[53]李琦.基于多源数据的交通状态监测与预测方法研究[D].长春：吉林大学，2013.
[54] Payne H J, Tignor S C. Freeway Incident Detection Algorithms Based onDecision Trees with States[J]. Transportation Research Record682, TRB,National Research Council, Washington D.C.,1978:30-37.
[55] Martin P, Perrin H, Hansen B. Incident detection algorithm evaluation [R].Technical Report UTL-0700-31, Utah Traffic Laboratory, July,2000.
[56]史新宏，蔡伯根．高速公路自动事件检测算法[J]．交通运输系统工程与信息，2011，37（8）：306-310．
[57] Dudek C L, Messer G M. Incident Detection on Urban Freeways[J].Transportation Research Record459, TRB, National Research Council,Washington D.C.,1974,(495):12-24.
[58] Cook A R, Cleveland D E. Detection of Freeway Capacdty Reducing Incidents byTraffic Stream Measurement[J]. Transportation Research Record, TRB,1974,(495):1-11.
[59] Martin P, Perrin H, Hansen B.Incident detection algorithm evaluation [R].Technical Report UTL-0700-31, Utah Traffic Laboratory, July,2000.
[60] Collins J F, Hopkins C M, Martin J A. Automatic Incident Detection. TRRLAlgorithms HIOCC and PATREG[R]. Supplementary Report526, Transport andRoad Research Laboratory, Crowthorne, England,1979.
[61] Ahmed S A, Cook A R. Application of Time-Series Analysis Techniques toFreeway Incident Detection[J]. Transportation Research Record682, TRB,National Research Council, Washington D.C.,1980,(841):19-21.
[62] Persaud B N, Hall F L, Hall L M. Congestion Identification Aspects of theMcMaster Incident Detection Algorithm[J]. Transportation Research Record,1990,(1287):167-175.
[63] Hsiao C H; Lin C T; Cassidy M. Application of Fuzzy Logic and NeuralNetworks to Automatically Detect Freeway Traffic Incidents[J]. Journal ofTransportation Engineering,1994,120(5):753-771.
[64] Ritchie S G; Cheu R L. Simulation of Freeway Incident Detection UsingArtificial Neural Networks[J]. Trasportation Research Part C,1993,1(3):313-331.
[65] Teng H, Qi Y. Application of wavelet technique to freeway incident detection[J].Transportation Research Part C: EmergingTechnologies,2003,11(3):289-308.
[66] Yuan F, Cheu R L. Incident detection using support vector machines[J].Transportation Research Part C: Emerging Technologies,2003,11(3):309-328.
[67]刘彬生.基于小波和傅立叶变换的道路交通量预测研究[D].哈尔滨：哈尔滨工业大学，2006.
[68]倪同和.道路交通规划关键指标预测方法研究[D].长春：吉林大学，2011.
[69]纪冬梅.基于组合模型的短时交通流预测方法[D].上海：华东理工大学，2012.
[70]喻丹.基于小波理论的短时交通流预测方法研究[D].长沙：长沙理工大学，2008.
[71]曹征.基于小波变换的交通流短时预测模型研究[D].北京：北京交通大学，2010.
[72]谭政.城市道路交通流预测及应用[D].广州：华南理工大学，2012.
[73]李存军.基于集成神经网络的城市道路交通流量融合预测研究[D].成都：西南交通大学，2004.
[74]李贞珍.神经网络在城市交通流预测模型中应用的研究[D].上海：上海交通大学，2007.
[75]王沥.基于改进神经网络的短时交通流预测研究[D].成都：电子科技大学，2012.
[76] Pauls J. Calculating evacuation times for tall buildings[J]. Fire Safety Journal,1987,12(3):213-236.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |