-
摘要: 鉴于采用数学建模的方法来划分服务水平的复杂性以及Kohonen网络的自组织学习能力和良好的鲁棒性, 提出了利用Kohonen网络对收费广场服务水平进行分类的方法, 并用该方法对一个含有不停车收费方式(Electronic Toll Collection, ETC) 与停车收费方式的收费广场服务水平进行了分类, 结果表明该方法是很有实用价值的。Abstract: In view of the difficulty and complexity of mathematical and physical method to classify the service levels and the good self-organization behavior and robustness of Kohonen neural network, this paper applied Kohonen neural network to classify the service levels for a certain toll plaza with ETC toll lanes and stop toll lanes. The results indicate that this method is potentially applicable in practice.
-
Key words:
- traffic engineering /
- toll plaza /
- service level /
- Kohonen neural network /
- ETC /
- classification
-
表 1 聚类中心服务水平分类结果
Table 1. Service level classification results of cluster center
服务水平 流率/veh·h-1 密度/veh· (km·lane)-1 F 2400 45.1 E 2075 34.9 D 1560 23.3 C 1140 15.5 B 770 10.7 A 420 6.7 表 2 聚类中心的流率通行能力比
Table 2. V/Caratio for cluster center
服务水平 流率V/veh·h-1 通行能力Ca/veh·h-1 流率/通行能力(V/Ca) F 2400 2330 1.03 E 2075 2330 0.89 D 1560 2330 0.67 C 1140 2330 0.49 B 770 2330 0.33 A 420 2330 0.18 -
[1] 刘伟铭. 高速公路收费系统理论与方法[M]. 北京: 人民交通出版社] 2000. [2] 美国交通研究委员会] 任福田. 道路通行能力手册(第三版) [M]. 北京: 中国建筑工业出版社] 1991. [3] 刘伟铭. 高速公路系统控制方法[M]. 北京: 人民交通出版社] 1998. [4] Lin Feng-Bor] Su Cheng-Wei. Level-of-service analysis of toll plazas on freeway main lines[J]. Journal of Transportation Engineering] 1994] 120(2): 246-263. doi: 10.1061/(ASCE)0733-947X(1994)120:2(246) [5] Lin F B] Lin M W. Modeling traffic delays at northern new york border crossings[J]. Journal of Transportation Engineering] 2001] 127(6): 540-545. doi: 10.1061/(ASCE)0733-947X(2001)127:6(540) [6] Bhargab Maitra] Sikdar P K] Dhingra S L. Modeling congestion on urban roads and assessing level of service[J]. Journal of Transportation Engineering] 1999] 125(6): 508-514. [7] Yang Hai] Qiao Feng-Xiang. Neural network approach to classification of traffic flow states[J]. Journal of Transportation Engineering] 1998] 124(6): 521-525. [8] 闻新. MATLAB神经网络应用设计[M]. 北京: 科学出版社] 2000.