基于模糊c均值的城市道路交通状态判别研究
|
摘要
随着经济的快速发展和城市化水平的不断提高,城市道路交通需求迅速增长与城市道路交通设施建设相对缓慢的矛盾越来越严重,从而引起了严重的交通问题,严重影响了道路交通的安全性和运行效率,因此,对作为交通诱导和交通控制基础的道路交通状态进行准确的判别是十分必要的。
本文以城市道路交通状态判别为研究对象,鉴于交通状态的模糊性和不确定性特征,本文选用模糊c均值算法对交通状态进行判别；在对城市道路交通状态判别指标进行研究的基础上,结合交通参数选取的原则,确定了本文进行交通状态判别的交通参数；为了尽量避免交通参数数据中噪声的干扰,从而得到更加准确的判别结果,本文对选取的交通参数数据进行了预处理；为了得到最优的聚类结果,本文对模糊c均值算法的模糊加权指数m和聚类类数c这两个参数进行了优选研究,从而得到基于选取路段交通参数数据的最优m、c值,同时,针对FCM算法收敛较慢的缺点,本文将HCM算法和FCM算法结合起来,很好的解决了这个问题；最后,本文选取一段快速路的数据进行了实例分析,得出了选取路段的交通状态,并对FCM算法的判别结果进行了评价。
With the rapid development of the economic and the continuous improvement of urbanization, the Contradiction between the transportation need and the traffic facilities becomes increasingly serious due to the fast increased need of road transportation and the relatively slow construction of road traffic facilities. Subsequently, transportation safety and efficiency have been going down gradually and leading to other traffic problems. As the basis of the traffic guidance and the traffic control, it is essential to get accurate traffic state identification.
In this paper, the urban road traffic state identification is researched. As the traffic state is fuzzy and uncertain, the fuzzy c-means algorithm is employed to identify the traffic state. Traffic state identification are determined on the basis of the research of the traffic state identification indexes according to the traffic parameter selection principle. The traffic data has been pretreated to decrease the interference of noise, and then the identification results are obtained more accurately. In order to obtain the best clustering results, fuzzy weighted index m and the number of clusters c are researched aimed to get the best optimized m and c values. To make up the convergence slower shortcoming of the FCM algorithm, the HCM algorithm is combined with the FCM algorithm, and it turns out to be a good solution to this problem. In the end, this solution has been put into practice, the traffic state of the selected urban freeway has been obtained accurately and its identification has been evaluated reasonably.
本文以城市道路交通状态判别为研究对象,鉴于交通状态的模糊性和不确定性特征,本文选用模糊c均值算法对交通状态进行判别；在对城市道路交通状态判别指标进行研究的基础上,结合交通参数选取的原则,确定了本文进行交通状态判别的交通参数；为了尽量避免交通参数数据中噪声的干扰,从而得到更加准确的判别结果,本文对选取的交通参数数据进行了预处理；为了得到最优的聚类结果,本文对模糊c均值算法的模糊加权指数m和聚类类数c这两个参数进行了优选研究,从而得到基于选取路段交通参数数据的最优m、c值,同时,针对FCM算法收敛较慢的缺点,本文将HCM算法和FCM算法结合起来,很好的解决了这个问题；最后,本文选取一段快速路的数据进行了实例分析,得出了选取路段的交通状态,并对FCM算法的判别结果进行了评价。
With the rapid development of the economic and the continuous improvement of urbanization, the Contradiction between the transportation need and the traffic facilities becomes increasingly serious due to the fast increased need of road transportation and the relatively slow construction of road traffic facilities. Subsequently, transportation safety and efficiency have been going down gradually and leading to other traffic problems. As the basis of the traffic guidance and the traffic control, it is essential to get accurate traffic state identification.
In this paper, the urban road traffic state identification is researched. As the traffic state is fuzzy and uncertain, the fuzzy c-means algorithm is employed to identify the traffic state. Traffic state identification are determined on the basis of the research of the traffic state identification indexes according to the traffic parameter selection principle. The traffic data has been pretreated to decrease the interference of noise, and then the identification results are obtained more accurately. In order to obtain the best clustering results, fuzzy weighted index m and the number of clusters c are researched aimed to get the best optimized m and c values. To make up the convergence slower shortcoming of the FCM algorithm, the HCM algorithm is combined with the FCM algorithm, and it turns out to be a good solution to this problem. In the end, this solution has been put into practice, the traffic state of the selected urban freeway has been obtained accurately and its identification has been evaluated reasonably.
引文
[1]李晓江等.中国城市交通发展战略[M].北京：中国建筑工业出版社,1997.
[2]宋宇.基于模糊综合评判的城市道路交通状态自动判别算法研究[D].昆明：云南大学,2010.
[3]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,2007.
[4]J.Wahle, O.Annen, Ch.Schuster, L.Neubert, M.Schreckenberg. A Dynamic Route Guidance System Based on Real Traffic Data[J]. European Journal of Operational Research,2001:302-308.
[5]Luk,JYK, Chung, ECS. CHARACTERIZATION OF INCIDENTS ON AN URBAN ARTERIAL ROAD[J]. Journal of Advanced Transportation,2001,35(1):67-92.
[6]Y.Li, M.Mcdonald. Motorway incident detection using probe vehicles[J]. Proceedings of the Institution of Civil Engineers,2004,(158):11-15.
[7]W.YIBINQM.PAPAGEORGIOU. Real-Time Freeway Traffic State Estimation Based on Extended Kalman Filter:A Case Study[J]. TRANSPORTATION SCIENCE,2007,41(2):167-181.
[8]Yaser, E.Hawas. A fuzzy-based system for incident detection in urban street networks. Transportation Research Part C, 2007,15:69-95.
[9]达庆东,姜学锋.基于流量-占有率检测数据的交通状态划分[J].道路交通与安全,2002,(3)：29-31.
[10]王建玲,蒋阳升.交通拥挤状态的识别与分析[J].系统工程,2006,24(10)：105-109.
[11]任其亮,谢小淞.基于遗传动态模糊聚类的道路交通状态判定方法[J].交通运输工程与信息学报,2007,5(3)：13-15,25.
[12]郝媛,徐天东,孙立军.基于模糊的城市快速路交通流状态判别[J].公路工程,2008,33(2)：94-99.
[13]任其亮,肖裕民.城市路网交通拥堵H-Fuzzy评判方法研究[J].重庆交通大学学报(自然科学版),2008,27(5)：763-766.
[14]杨祖元,黄席樾,杜长海,唐明霞.基于FFCM聚类的城市交通拥堵判别研究[J].计算机应用研究,2008,25(9)：2768-2770.
[15]林明文.基于实时交通流检测的道路交通状态判定系统[D].大连：大连理工大学,2009.
[16]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,2001,1(4)：306-310.
[17]王东.城市交通拥堵状态自动判别方法研究[D].贵阳：贵州大学,2008.
[18]戢晓峰,刘澜.基于交通信息提取的区域交通状态判别方法[J]三：峡大学学报(自然科学版),2009,31(1)：94-97.
[19]徐磊.城市道路交通状态判别与预测系统设计与关键技术研究[D].长春：吉林大学,2010.
[20]王吟松,王嘉文,张建跃,董秀军,杨晓光.基于浮动车车速数据的交通状态实时判别系统和方法[J].交通信息与安全,2011,29(6)：97-102.
[21]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.
[22]温娟,贺国光.基于粗集理论预处理数据的神经网络交通事件自动检测算法[J].交通运 输系统工程与信息,2004,4(4)：54-59.
[23]聂佩林,陈晓翔,余锡伟,戴秀斌.基于代价敏感神经网络的交通状态判别[J].公路交通科技(应用技术版),2011,7(3).
[24]巫威眺,靳文舟,林培群.基于BP神经网络的道路交通状态判别方法研究[J].交通信息与安全,2011,4(29)：71-79.
[25]马世勇,郑应文,柯栋忠.城市交通拥堵自动判别方法研究[J].福建电脑,2007,(1)：7-8.
[26]李江.交通工程学[M].北京：人民交通出版社,2007.
[27]D.Whitey. Flexible scheduling in a machining center through genetic algorithms[J]. Computers and Industrial Engineering,199630(4):931-940.
[28]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.
[29]欧冬秀.交通信息技术[M].上海：同济大学出版社,2007.
[30]徐晓慧,王德章.道路交通控制教程[M].北京：中国人民公安大学出版社,2005.
[31]Shawn Turner, Luke Albert, Byron Gajewski etc. Archived Its Data Quality: Preliminary Analysis Of San Antonio Tranguide Data[A]. The 79th annual meeting transportation research board Washington D.C,2000.
[32]Rod E.Turochy, Brian L.Smith. A New Procedure for Detector Data Screening In Traffic Management Systems.The 80th annual meeting Transportation Research Board. Washington D.C,2001:1-19.
[33]L.N. Jacobson, N.L. Nihan, J.D. Bender. Detecting Erroneous Loop Detector Data in a Freeway Traffic Management System. Transportation Research Record. National Research Council. Washington D.C,1990.
[34]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,2003.36(7)：78-83.
[35]储浩,杨晓光,吴志周.动态交通数据的预处理方法及应用[A].第一届中国智能交通年会论文集.2005.
[36]秦玲,郭艳梅,吴鹏,冯剑,桑丽.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)：39-41.
[37]姜桂艳,江龙晖,张晓东,王江锋.动态交通数据故障识别与修复方法[J].交通运输工程学报,2004,4(1)：121-125.
[38]韦清波.基于交通流预测的高速公路交通状态判别方法研究[D].广州：中山大学,2010.
[39]杨兆升.基础交通信息融合技术及其应用[M].北京：中国铁道出版社,2005.
[40]GA115-1995.道路交通阻塞及评价方法[S].
[41]2006年北京市快速路及主次干道交通状况调查报告[R].北京：北京市交通发展研究中心,2006.
[42]关伟,何蜀燕.基于统计特性的城市快速路交通流状态划分[J].交通运输系统工程与信息,2007,7(5)：42-50.
[43]翟雅峤,翁剑成,荣建,刘小明.多源数据融合的区间车辆速度预测算法研究[J].交通信息与安全,2009,27(3)：74-77.
[44]龙小强,谭云龙.基于模糊综合评价的城市道路交通拥堵评价研究[J].交通工程与安全,2011,(11)：114-117.
[45]成卫,李江.模糊聚类法在基于交通冲突技术的交叉口安全评价中的应用[J].交通运输系统工程与信息,2004,4(2)：48-55.
[46]袁红梅.模糊c均值算法的研究[D].银川：宁夏大学,2010.
[47]Kamel M S, Selim S Z. New algorithms for solving the fuzzy c-means Clustering problem[J]. Pattern Recognition,1994,27:421.
[48]XIE X, BENI G A. validity measure for Fuzzy clustering[J]. Transactions on Patten Analysis and Machine Intelligence,1991,13(8):841-847.
[49]Pal N R, Pal K, Keller J M, Bezdek J C. A Possibilistic Fuzzy c-Means Clustering Algorithm[J].2005 13(4),517-530.
[50]Hathaway R J, Bezdek J C. Extending fuzzy and probabilistic clustering to very large data sets[J]. Computational Statistics & Data Analysis,2006,51(1):215-234.
[51]Bezdek J C. Patten recognition with fuzzy objective function algorithms[M]. New York: Plenum Press,1981.
[52]宫改云,高新波,伍忠东.FCM聚类算法中模糊加权指数m的优选方法[J].模糊系统与数学,2005,19(1)：143-148.
[53]Cheung Y S, Chan K P. Modified fuzzy ISODATA for the classification of handwriting Chinese characters[A]. Proc Int Conf Chinese Comput,Singapore,1986:361-364.
[54]Bezdek J C,Hathaway R J, et al.Convergence theory for fuzzy c-means:Counter-examples and repairs [J]. IEEE Trans. Syst. Man Cybern,1987,17(5):873-877.
[55]Pal N R, Bezdek J C. On cluster validity for the fuzzy c-means model[J]. IEEE Trans. Fuzzy Systems,1995,3(3)370-379.
[56]高新波,裴继红,谢维信.模糊c均值聚类算法中加权指数m的研究[J].电子学报,2004,28(4)：14.
[57]刘宜平,沈毅,童树鸿,等.一种参数在线优化的FCM改进算法[J].电机与控制学报,2000,3(4)：164-167.
[58]于剑.论模糊C均值算法的模糊指标[J].计算机学报,2003,26(8)：968-973.
[59]姜琴,甘海涛.FCM算法中参数确定方法的探讨[J].武汉工业学院学报,2009,28(1)42-44,63.
[60]姜桂艳,郭海锋,吴超腾.基于感应线圈数据的城市道路交通状态判别方法[J].吉林大学学报(工业版),2008,38：37-42.
[61]GBJ124-88,道路工程术语标准[S].
[62]梅君智.BP神经网络的结构优化及应用[D].广州：中山大学,2010.
[2]宋宇.基于模糊综合评判的城市道路交通状态自动判别算法研究[D].昆明：云南大学,2010.
[3]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,2007.
[4]J.Wahle, O.Annen, Ch.Schuster, L.Neubert, M.Schreckenberg. A Dynamic Route Guidance System Based on Real Traffic Data[J]. European Journal of Operational Research,2001:302-308.
[5]Luk,JYK, Chung, ECS. CHARACTERIZATION OF INCIDENTS ON AN URBAN ARTERIAL ROAD[J]. Journal of Advanced Transportation,2001,35(1):67-92.
[6]Y.Li, M.Mcdonald. Motorway incident detection using probe vehicles[J]. Proceedings of the Institution of Civil Engineers,2004,(158):11-15.
[7]W.YIBINQM.PAPAGEORGIOU. Real-Time Freeway Traffic State Estimation Based on Extended Kalman Filter:A Case Study[J]. TRANSPORTATION SCIENCE,2007,41(2):167-181.
[8]Yaser, E.Hawas. A fuzzy-based system for incident detection in urban street networks. Transportation Research Part C, 2007,15:69-95.
[9]达庆东,姜学锋.基于流量-占有率检测数据的交通状态划分[J].道路交通与安全,2002,(3)：29-31.
[10]王建玲,蒋阳升.交通拥挤状态的识别与分析[J].系统工程,2006,24(10)：105-109.
[11]任其亮,谢小淞.基于遗传动态模糊聚类的道路交通状态判定方法[J].交通运输工程与信息学报,2007,5(3)：13-15,25.
[12]郝媛,徐天东,孙立军.基于模糊的城市快速路交通流状态判别[J].公路工程,2008,33(2)：94-99.
[13]任其亮,肖裕民.城市路网交通拥堵H-Fuzzy评判方法研究[J].重庆交通大学学报(自然科学版),2008,27(5)：763-766.
[14]杨祖元,黄席樾,杜长海,唐明霞.基于FFCM聚类的城市交通拥堵判别研究[J].计算机应用研究,2008,25(9)：2768-2770.
[15]林明文.基于实时交通流检测的道路交通状态判定系统[D].大连：大连理工大学,2009.
[16]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,2001,1(4)：306-310.
[17]王东.城市交通拥堵状态自动判别方法研究[D].贵阳：贵州大学,2008.
[18]戢晓峰,刘澜.基于交通信息提取的区域交通状态判别方法[J]三：峡大学学报(自然科学版),2009,31(1)：94-97.
[19]徐磊.城市道路交通状态判别与预测系统设计与关键技术研究[D].长春：吉林大学,2010.
[20]王吟松,王嘉文,张建跃,董秀军,杨晓光.基于浮动车车速数据的交通状态实时判别系统和方法[J].交通信息与安全,2011,29(6)：97-102.
[21]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.
[22]温娟,贺国光.基于粗集理论预处理数据的神经网络交通事件自动检测算法[J].交通运 输系统工程与信息,2004,4(4)：54-59.
[23]聂佩林,陈晓翔,余锡伟,戴秀斌.基于代价敏感神经网络的交通状态判别[J].公路交通科技(应用技术版),2011,7(3).
[24]巫威眺,靳文舟,林培群.基于BP神经网络的道路交通状态判别方法研究[J].交通信息与安全,2011,4(29)：71-79.
[25]马世勇,郑应文,柯栋忠.城市交通拥堵自动判别方法研究[J].福建电脑,2007,(1)：7-8.
[26]李江.交通工程学[M].北京：人民交通出版社,2007.
[27]D.Whitey. Flexible scheduling in a machining center through genetic algorithms[J]. Computers and Industrial Engineering,199630(4):931-940.
[28]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.
[29]欧冬秀.交通信息技术[M].上海：同济大学出版社,2007.
[30]徐晓慧,王德章.道路交通控制教程[M].北京：中国人民公安大学出版社,2005.
[31]Shawn Turner, Luke Albert, Byron Gajewski etc. Archived Its Data Quality: Preliminary Analysis Of San Antonio Tranguide Data[A]. The 79th annual meeting transportation research board Washington D.C,2000.
[32]Rod E.Turochy, Brian L.Smith. A New Procedure for Detector Data Screening In Traffic Management Systems.The 80th annual meeting Transportation Research Board. Washington D.C,2001:1-19.
[33]L.N. Jacobson, N.L. Nihan, J.D. Bender. Detecting Erroneous Loop Detector Data in a Freeway Traffic Management System. Transportation Research Record. National Research Council. Washington D.C,1990.
[34]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,2003.36(7)：78-83.
[35]储浩,杨晓光,吴志周.动态交通数据的预处理方法及应用[A].第一届中国智能交通年会论文集.2005.
[36]秦玲,郭艳梅,吴鹏,冯剑,桑丽.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)：39-41.
[37]姜桂艳,江龙晖,张晓东,王江锋.动态交通数据故障识别与修复方法[J].交通运输工程学报,2004,4(1)：121-125.
[38]韦清波.基于交通流预测的高速公路交通状态判别方法研究[D].广州：中山大学,2010.
[39]杨兆升.基础交通信息融合技术及其应用[M].北京：中国铁道出版社,2005.
[40]GA115-1995.道路交通阻塞及评价方法[S].
[41]2006年北京市快速路及主次干道交通状况调查报告[R].北京：北京市交通发展研究中心,2006.
[42]关伟,何蜀燕.基于统计特性的城市快速路交通流状态划分[J].交通运输系统工程与信息,2007,7(5)：42-50.
[43]翟雅峤,翁剑成,荣建,刘小明.多源数据融合的区间车辆速度预测算法研究[J].交通信息与安全,2009,27(3)：74-77.
[44]龙小强,谭云龙.基于模糊综合评价的城市道路交通拥堵评价研究[J].交通工程与安全,2011,(11)：114-117.
[45]成卫,李江.模糊聚类法在基于交通冲突技术的交叉口安全评价中的应用[J].交通运输系统工程与信息,2004,4(2)：48-55.
[46]袁红梅.模糊c均值算法的研究[D].银川：宁夏大学,2010.
[47]Kamel M S, Selim S Z. New algorithms for solving the fuzzy c-means Clustering problem[J]. Pattern Recognition,1994,27:421.
[48]XIE X, BENI G A. validity measure for Fuzzy clustering[J]. Transactions on Patten Analysis and Machine Intelligence,1991,13(8):841-847.
[49]Pal N R, Pal K, Keller J M, Bezdek J C. A Possibilistic Fuzzy c-Means Clustering Algorithm[J].2005 13(4),517-530.
[50]Hathaway R J, Bezdek J C. Extending fuzzy and probabilistic clustering to very large data sets[J]. Computational Statistics & Data Analysis,2006,51(1):215-234.
[51]Bezdek J C. Patten recognition with fuzzy objective function algorithms[M]. New York: Plenum Press,1981.
[52]宫改云,高新波,伍忠东.FCM聚类算法中模糊加权指数m的优选方法[J].模糊系统与数学,2005,19(1)：143-148.
[53]Cheung Y S, Chan K P. Modified fuzzy ISODATA for the classification of handwriting Chinese characters[A]. Proc Int Conf Chinese Comput,Singapore,1986:361-364.
[54]Bezdek J C,Hathaway R J, et al.Convergence theory for fuzzy c-means:Counter-examples and repairs [J]. IEEE Trans. Syst. Man Cybern,1987,17(5):873-877.
[55]Pal N R, Bezdek J C. On cluster validity for the fuzzy c-means model[J]. IEEE Trans. Fuzzy Systems,1995,3(3)370-379.
[56]高新波,裴继红,谢维信.模糊c均值聚类算法中加权指数m的研究[J].电子学报,2004,28(4)：14.
[57]刘宜平,沈毅,童树鸿,等.一种参数在线优化的FCM改进算法[J].电机与控制学报,2000,3(4)：164-167.
[58]于剑.论模糊C均值算法的模糊指标[J].计算机学报,2003,26(8)：968-973.
[59]姜琴,甘海涛.FCM算法中参数确定方法的探讨[J].武汉工业学院学报,2009,28(1)42-44,63.
[60]姜桂艳,郭海锋,吴超腾.基于感应线圈数据的城市道路交通状态判别方法[J].吉林大学学报(工业版),2008,38：37-42.
[61]GBJ124-88,道路工程术语标准[S].
[62]梅君智.BP神经网络的结构优化及应用[D].广州：中山大学,2010.