In order to identify the expressway traffic state effectively and improve the service level of the road network, the traffic data of expressway was analyzed based on the improved FCM (Fuzzy C-Means) clustering algorithm. Firstly, entropy weight method was used to determine the weights of four traffic state classification indexes, including traffic flow, space occupancy, average speed and road network ample degree, meanwhile different weighting coefficient was assigned to each sample. Secondly, the calculation of sample weights was incorporated into the iterative process to identify the expressway traffic state. Finally, the objective function value, iteration times and running time of the improved FCM algorithm and the traditional FCM algorithm were compared. The results show that compared with the traditional FCM algorithm, the improved FCM algorithm has smaller objective function value, fewer iterations, shorter running time and better adaptability in data; the clustering results obtained by the improved FCM algorithm can reflect the changes of traffic data accurately and comprehensively, and can identify road traffic states effectively.
余 庆，胡 尧
. Identification on Traffic State of Expressway Based on Improved FCM Clustering Algorithm[J]. Transport Research, 2021
, 7(2)
: 47
-54
.
DOI: 10.16503/j.cnki.2095-9931.2021.02.006
[1] YU X H, XIONG S W, HE Y, et al. Research on campus traffic congestion detection using BP neural network and Markov model[J]. Journal of Information Security and Applications, 2016, 31: 54-60.
[2] SUN Q X, SUN Y X, SUN L, et al. Research on traffic congestion characteristics of city business circles based on TPI data: the case of Qingdao, China[J]. Physica A: Statistical Mechanics and its Applications, 2019, 534(3): 122214.
[3] 郭海涛. 基于图像识别技术的区域交通拥堵状态判别研究[J]. 信息记录材料，2019，20（1）：74-76.
[4] 彭博，唐聚，蔡晓禹，等. 基于3DCNN-DNN的高空视频交通状态预测[J]. 交通运输系统工程与信息，2020，20（3）：39-46.
[5] RICARDO G, MARIA L L, MARIA T S. An approach to dynamical classification of daily traffic patterns[J]. Computer-Aided Civil and Infrastructure Engineering, 2017, 32(3): 191-212.
[6] BAE B, LIU Y D, HAN L D, et al. Spatio-temporal traffic queue detection for uninterrupted flows[J]. Transportation Research Part B: Methodological, 2019, 129: 20-34.
[7] 陈忠辉，凌献尧，冯心欣，等. 基于模糊C均值聚类和随机森林的短时交通状态预测方法[J]. 电子与信息学报，2018，40（8）：1879-1886.
[8] 陈钊正，吴聪. 多变量聚类分析的高速公路交通流状态实时评估[J]. 交通运输系统工程与信息，2018，18（3）：225-233.
[9] 张亮亮，贾元华，牛忠海，等. 交通状态划分的参数权重聚类方法研究[J]. 交通运输系统工程与信息，2014，14（6）：147-151.
[10] 曹洁，张丽君，侯亮，等. 基于信息熵加权的FCM交通状态识别研究[J]. 计算机应用与软件，2018，35（10）：68-73.
[11] 王宇俊，田锋，叶道均，等. 改进FCM的交通状态判别算法[C]// 交叉创新与转型重构——2017年中国城市交通规划年会论文集. 北京：中国建筑工业出版社，2017.
[12] 于泉，孙瑶. 基于组合赋权法的城市交叉口交通状态评价[J]. 交通运输研究，2018，4（2）：23-29.
[13] CALTRANS. Performance Measurement System
(PeMS)[DB/OL]. (2020-07-07)[2020-08-05].
[14] 孙晓亮. 城市道路交通状态评价和预测方法及应用研究[D]. 北京：北京交通大学，2013.
[15] CHENG Z Y, WANG W, LU J, et al. Classifying the traffic state of urban expressways: A machine-learning approach[J]. Transportation Research Part A: Policy and Practice, 2020, 137: 411-428.
[16] 章永来，周耀鉴. 聚类算法综述[J]. 计算机应用，2019，39（7）：1869-1882.
[17] 何兴高，李蝉娟，王瑞锦，等. 基于信息熵的高维稀疏大数据降维算法研究[J]. 电子科技大学学报，2018，47（2）：235-241.
[18] LIN J X, WU L P, CHEN R Q, et al. Double-weighted fuzzy clustering with samples and generalized entropy features[J]. Concurrency and Computation: Practice and Experience, 2020, 33(8): 1-15.
