基于(SAGA-FCM)-PNN的交通状态判别方法研究
|
摘要
为了提高城市道路交通状态判别的正确性与稳定性,研究了一种基于遗传模拟退火算法改进的FCM算法与概率神经网络(PNN)结合的短时交通流状态判别方法。针对传统FCM算法会收敛到局部最优解的问题,利用遗传模拟退火算法对其进行改进,优化算法初始聚类中心;将已分类的数据分为训练集与测试集对概率神经网络(PNN)模型进行训练与测试,通过对径向基函数的扩展速度的优化提高PNN算法的准确性;并利用厦门市城市道路地磁检测数据对模型进行实例验证及性能分析。结果表明,文中方法能够有效的实现交通状态的判别,且能够得到全局最优解;同竞争神经网络模型、GRNN模型、SVM模型相比,文中模型的交通状态判别正确率分别提高2.1%,4.5%,2.7%,且具有更好的稳定性。
In order to improve accuracy and stability of discrimination for traffic state, a method to discriminate short-time traffic flow is studied based on SAGA-FCM and probabilistic neural network(PNN). SAGA can solve problems that traditional FCM converges to the locally optimal solution, and improves the initial clustering center. Classified data is divided into training set and test set to trained and tested PNN. The accuracy of PNN is improved by optimizing expansion speed of a radial basis function. A case study is carried out by using geomagnetic detection data of Xiamen urban roads to verify and analyze the model. The result shows that the proposed method can effectively discriminate the traffic state, and obtain the global optimal solution. Compared with competitive neural network model, GRNN, and SVM, the accuracy of this model is improved by 2.1%, 4.5%, and 2.7%, respectively, with better stability.
In order to improve accuracy and stability of discrimination for traffic state, a method to discriminate short-time traffic flow is studied based on SAGA-FCM and probabilistic neural network(PNN). SAGA can solve problems that traditional FCM converges to the locally optimal solution, and improves the initial clustering center. Classified data is divided into training set and test set to trained and tested PNN. The accuracy of PNN is improved by optimizing expansion speed of a radial basis function. A case study is carried out by using geomagnetic detection data of Xiamen urban roads to verify and analyze the model. The result shows that the proposed method can effectively discriminate the traffic state, and obtain the global optimal solution. Compared with competitive neural network model, GRNN, and SVM, the accuracy of this model is improved by 2.1%, 4.5%, and 2.7%, respectively, with better stability.
引文
[1] 商强,林赐云,杨兆升.基于谱聚类与RS-KNN的城市快速路交通状态判别[J].华南理工大学学报(自然科学版),2017,45(6):52-58.SHANG Qiang,LIN Ciyun,YANG Zhaosheng.Traffic state identification for urban expressway based on spectral clustering and RS-KNN[J].Journal of South China University of Technology (Natural Science Edition),2017,45(6):52-58.(in Chinese)
[2] 杨庆芳,马明辉,梁士栋,等.基于收费数据的高速公路交通状态判别方法[J].华南理工大学学报(自然科学版),2014,42(12):51-57,76.YANG Qingfang,MA Minghui,LIANG Shidong,et al.Freeway traffic state identification based on toll data[J].Journal of South China University of Technology (Natural Science Edition),2014,42(12):51-57,76.(in Chinese)
[3] 杨祖元,徐姣,罗兵.基于SFLA-FCM聚类的城市交通状态判别研究[J].计算机应用研究,2010,27(5):1743-1745.YANG Zuyuan,XU Jiao,LUO Bing.Urban Traffic condition discrimination based on SFLA-FCM clustering[J] .Application Research of Computers.2010,42(12):1743-1745.(in Chinese)
[4] 邴其春,龚勃文,杨兆升,等.基于投影寻踪动态聚类的快速路交通状态判别[J].西南交通大学学报,2015,50(6):1164-1169.BING Qichun,GONG Bowen,YANG Zhaosheng.Traffic state identification for urban expressway basedon projection pursuit dynamic cluster model [J].Journal of Southwest Jiaotong University.2015,50(6):1164-1169.(in Chinese)
[5] MONTAZE I-GH M,FOTOUHI A.Traffic condition re-cognition using the k-means clustering method[J].Scientia Iranica,2011,18(4):930-937.
[6] 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.CAO Jie,ZHANG Lijun,HOU Liang,et al.Research on FCM traffic status recognition based on entropy weighting [J].Computer application and software,2008,35(10):68-73.(in Chinese)
[7] 陈征,刘钊,王静,等.基于RFID和视频监测的城市交通运行状态模糊判别[J].交通信息与安全,2018,36(2):68-75.CHEN Zheng,LIU Zhao,WANG Jing,et al.A fuzzy merhod for identifying urban traffic state using rfid and video monitoring system[J].Journal of Transport Information and Safety,2018,36(2):68-75.(in Chinese)
[8] NIU S,LIU H.Probabilistic neural networks for theidentification of traffic state[C].14th International IEEE Conference on Intelligent Transportation Systems(ITSC).Washington,D.C.:IEEE,2011:754-759.
[9] 董春娇,邵春福,熊志华.基于优化SVM的城市快速路网交通流状态判别[J].北京交通大学学报,2011,35(6):13-16,22.DONG Chunjiao,SHAO Chunfu,XIONG Zhihua.Identification of traffic states with optimized SVM method on urban expressway network[J].Journal of Beijing Jiaotong University,2011,35(6):13-16,22.(in Chinese)
[10] 聂佩林,陈晓翔,佘锡伟,等.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):220-223.NIE Peilin,CHEN Xiaoxiang,SHE Xiwei,et al.Traffic state discrimination based on cost-sensitive neural network[J].Journal of Highway and Transportation Research and Development(Applied Technology Edition),2011,7(3):220-223.(in Chinese)
[11] 启顺.基于PNN的山地快速路交通状态判别方法及应用[C].2016中国城市规划年会,沈阳,中国:中国城市规划学会,2016.WU Qishun.PNN-based distinguishing method for mountain expressway traffic status and application[C].Annual Nation Planning Conference 2016,Shenyang,China:Urban Planning Society of China,2016.(in Chinese)
[12] 高林.随机森林算法在交通状态判别中的应用[C].第27届中国过程控制会议(CPCC2016),兰州,中国:中国自动化学会过程控制专业委员会,2016.GAO Lin.The application of the random forest algorithm in traffic status discriminant[C].The 27th China process control conference (CPCC2016),Lanzhou,China:Technical Committee on Process Control,Chinese Association of Automation,2016.(in Chinese)
[13] XIA J,CHEN M.A nested clustering technique for freeway operating condition classification[J].Computer-Aided Civil and Infrastructure Engineering,2007,22(6):430-437.
[14] DENG C,WANG F,SHI H,et al.Real-time freeway traffic state estimation based on cluster analysis and multi class support vector machine[C].International Workshop on Intelligent Systems and Applications(ISA),Wuhan:IEEE,2009.
[15] 孙壮文.基于KNN-HA和KNN-RBF相融合的交通状态预测[D].西安:长安大学,2017.SUN Zhuangwen.Traffic state prediction based on KNN-HA and KNN-RBF fusion[D].Xi′an:Chang′an University,2017.(in Chinese)
[16] 刘擎超,蔡英凤,江浩斌.基于强空间集成的交通状态判别方法[J].计算机工程,2017,43(11):297-302.LIU Qingchao,CAI Yingfeng,JIANG Haobin.Automatic Identification method based on strong space integration[J],Computer Engineering,2017,43(11):297-302.(in Chinese)
[17] 王占武.径向基函数扩展速度因子对预测结果的影响[J].智能城市,2016,2(1):152-154.WANG Zhanwu.Influence of spread velocity factor for radial basis function on the prediction results[J].Smart City,2016,2(1):152-154.(in Chinese)
[18] VLAHOGIANNI E I,KARLAFTIS M G,GOLIAS J C.Optimized and meta-optimized neural networks for short-term traffic flow prediction:A genetic approach[J].Transportation Research Prat C:Emerging Technologies,2005,13(3):211-234.
[19] ZHONG M G,SHARMA S,LINGRAS P.Genetically designed models for accurate imputation of missing traffic counts[J].Journal of the Transportation Research Board,2004,1879(1):71-79.
[2] 杨庆芳,马明辉,梁士栋,等.基于收费数据的高速公路交通状态判别方法[J].华南理工大学学报(自然科学版),2014,42(12):51-57,76.YANG Qingfang,MA Minghui,LIANG Shidong,et al.Freeway traffic state identification based on toll data[J].Journal of South China University of Technology (Natural Science Edition),2014,42(12):51-57,76.(in Chinese)
[3] 杨祖元,徐姣,罗兵.基于SFLA-FCM聚类的城市交通状态判别研究[J].计算机应用研究,2010,27(5):1743-1745.YANG Zuyuan,XU Jiao,LUO Bing.Urban Traffic condition discrimination based on SFLA-FCM clustering[J] .Application Research of Computers.2010,42(12):1743-1745.(in Chinese)
[4] 邴其春,龚勃文,杨兆升,等.基于投影寻踪动态聚类的快速路交通状态判别[J].西南交通大学学报,2015,50(6):1164-1169.BING Qichun,GONG Bowen,YANG Zhaosheng.Traffic state identification for urban expressway basedon projection pursuit dynamic cluster model [J].Journal of Southwest Jiaotong University.2015,50(6):1164-1169.(in Chinese)
[5] MONTAZE I-GH M,FOTOUHI A.Traffic condition re-cognition using the k-means clustering method[J].Scientia Iranica,2011,18(4):930-937.
[6] 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.CAO Jie,ZHANG Lijun,HOU Liang,et al.Research on FCM traffic status recognition based on entropy weighting [J].Computer application and software,2008,35(10):68-73.(in Chinese)
[7] 陈征,刘钊,王静,等.基于RFID和视频监测的城市交通运行状态模糊判别[J].交通信息与安全,2018,36(2):68-75.CHEN Zheng,LIU Zhao,WANG Jing,et al.A fuzzy merhod for identifying urban traffic state using rfid and video monitoring system[J].Journal of Transport Information and Safety,2018,36(2):68-75.(in Chinese)
[8] NIU S,LIU H.Probabilistic neural networks for theidentification of traffic state[C].14th International IEEE Conference on Intelligent Transportation Systems(ITSC).Washington,D.C.:IEEE,2011:754-759.
[9] 董春娇,邵春福,熊志华.基于优化SVM的城市快速路网交通流状态判别[J].北京交通大学学报,2011,35(6):13-16,22.DONG Chunjiao,SHAO Chunfu,XIONG Zhihua.Identification of traffic states with optimized SVM method on urban expressway network[J].Journal of Beijing Jiaotong University,2011,35(6):13-16,22.(in Chinese)
[10] 聂佩林,陈晓翔,佘锡伟,等.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):220-223.NIE Peilin,CHEN Xiaoxiang,SHE Xiwei,et al.Traffic state discrimination based on cost-sensitive neural network[J].Journal of Highway and Transportation Research and Development(Applied Technology Edition),2011,7(3):220-223.(in Chinese)
[11] 启顺.基于PNN的山地快速路交通状态判别方法及应用[C].2016中国城市规划年会,沈阳,中国:中国城市规划学会,2016.WU Qishun.PNN-based distinguishing method for mountain expressway traffic status and application[C].Annual Nation Planning Conference 2016,Shenyang,China:Urban Planning Society of China,2016.(in Chinese)
[12] 高林.随机森林算法在交通状态判别中的应用[C].第27届中国过程控制会议(CPCC2016),兰州,中国:中国自动化学会过程控制专业委员会,2016.GAO Lin.The application of the random forest algorithm in traffic status discriminant[C].The 27th China process control conference (CPCC2016),Lanzhou,China:Technical Committee on Process Control,Chinese Association of Automation,2016.(in Chinese)
[13] XIA J,CHEN M.A nested clustering technique for freeway operating condition classification[J].Computer-Aided Civil and Infrastructure Engineering,2007,22(6):430-437.
[14] DENG C,WANG F,SHI H,et al.Real-time freeway traffic state estimation based on cluster analysis and multi class support vector machine[C].International Workshop on Intelligent Systems and Applications(ISA),Wuhan:IEEE,2009.
[15] 孙壮文.基于KNN-HA和KNN-RBF相融合的交通状态预测[D].西安:长安大学,2017.SUN Zhuangwen.Traffic state prediction based on KNN-HA and KNN-RBF fusion[D].Xi′an:Chang′an University,2017.(in Chinese)
[16] 刘擎超,蔡英凤,江浩斌.基于强空间集成的交通状态判别方法[J].计算机工程,2017,43(11):297-302.LIU Qingchao,CAI Yingfeng,JIANG Haobin.Automatic Identification method based on strong space integration[J],Computer Engineering,2017,43(11):297-302.(in Chinese)
[17] 王占武.径向基函数扩展速度因子对预测结果的影响[J].智能城市,2016,2(1):152-154.WANG Zhanwu.Influence of spread velocity factor for radial basis function on the prediction results[J].Smart City,2016,2(1):152-154.(in Chinese)
[18] VLAHOGIANNI E I,KARLAFTIS M G,GOLIAS J C.Optimized and meta-optimized neural networks for short-term traffic flow prediction:A genetic approach[J].Transportation Research Prat C:Emerging Technologies,2005,13(3):211-234.
[19] ZHONG M G,SHARMA S,LINGRAS P.Genetically designed models for accurate imputation of missing traffic counts[J].Journal of the Transportation Research Board,2004,1879(1):71-79.