基于神经网络及关联性修正的交通异常预测研究
|
摘要
为实现城市交通异常管理的主动式响应,给交通异常处置争取更多的时间,减少交通异常对城市路网的影响,提出了一种基于神经网络及关联性修正的交通异常预测方法。基于历史异常数据构建交通异常数据库,并定义了预测模型中的主要参数;构建了基于改进神经网络算法的交通异常预测模型,在此基础上,创新性地提出了结合不同单元区域背景概率及交通异常相关关系挖掘的预测修正算法,对预测结果进行关联性修正以得到最终更加准确的预测结果,大幅提升了模型的预测精度。应用哈尔滨市30 d实例数据训练了所提出的交通异常预测模型,用15 d数据进行了验证,结果表明经过关联性修正的预测模型成功预测次数明显增加,相较于传统方法,预测成功率提升了31.46%,皮尔逊检验值均大于1.642,预测的结果的可信度大于80%的置信水平。
In order to achieve active response and strive for more disposal time of abnormal traffic in urban management, and reduce impacts of abnormal traffic on urban road networks, a model of forecasting abnormal traffic based on neural network and relevant correction. A database of abnormal traffic is established based on historical abnormal data, and defines main parameters of the model. Then, a method based on improved neural network algorithm is developed. This paper innovatively proposes a correction algorithm for forecasting with a combination of background probability of different unit regions and the mining results of abnormal traffic correlation. Accuracy of forecast is greatly improved after relevant correction. A case study with 30-day training data and 15-day test data collected in Harbin is applied to the model. The results show that compared with traditional methods, success rate of forecast improves by 31.46% when using the new method. The values of Pearson test are all above than 1.642, and confidence level of reliability is more than 80%.
In order to achieve active response and strive for more disposal time of abnormal traffic in urban management, and reduce impacts of abnormal traffic on urban road networks, a model of forecasting abnormal traffic based on neural network and relevant correction. A database of abnormal traffic is established based on historical abnormal data, and defines main parameters of the model. Then, a method based on improved neural network algorithm is developed. This paper innovatively proposes a correction algorithm for forecasting with a combination of background probability of different unit regions and the mining results of abnormal traffic correlation. Accuracy of forecast is greatly improved after relevant correction. A case study with 30-day training data and 15-day test data collected in Harbin is applied to the model. The results show that compared with traditional methods, success rate of forecast improves by 31.46% when using the new method. The values of Pearson test are all above than 1.642, and confidence level of reliability is more than 80%.
引文
[1] 韩慧龙.基于数据挖掘的交通流机理分析[D].广州:华南理工大学,2014.HAN Huilong.Analysis of traffic mechanism based on data mining [D].Guangzhou:South China University of Technology,2014.(in Chinese)
[2] 马玄.城市道路交通流预测与状态判别关键技术研究[D].广州:华南理工大学,2014.MA Xuan.Research on the key technologies of urban road traffic flow prediction and state discrimination[D].Guangzhou:South China University of Technology,2014.(in Chinese)
[3] 谷远利,张源,芮小平.基于免疫算法优化LSSVM的短时交通流预测研究[J/OL].(2018-09-18)[2018-11-16].https://doi.org/10.13229/j.cnki.jdxbgxb20180540.GU Yuanli,ZHANG Yuan,RUI Xiaoping.A short-term traffic flow prediction based on lssvm optimized by immune algorithm [J/OL].(2018-09-18)[2018-11-16].https://doi.org/10.13229/j.cnki.jdxbgxb 20180540.(in Chinese)
[4] 王体迎,时鹏超,刘蒋琼,等.基于门限递归单元循环神经网络的交通流预测方法研究[J].重庆交通大学学报(自然科学版),2018,37(11):76-82.WANG Tiying,SHI Pengchao,LIU Jiangqiong,et al.Research on traffic flow prediction method based on gated recurrent unit recurrent neural network [J].Journal of Chongqing Jiaotong University (Natural Science Edition),2018,37(11)76-62.(in Chinese)
[5] 韩春颖,周亚同,常和玲,等.基于稀疏高斯过程混合模型的短时交通流预测[J].交通信息与安全,2019,37(1):121-127.HAN Chunying,ZHOU Yatong,CHANG He Ling,CHI Yue,HE Jingfei.A short-term forecasting of traffic flow based on a mixture model of sparse gaussian process [J].Journal of Transport Safety,2019,37(1):121-127.(in Chinese)
[6] 冯微,陈红,张兆津,等.基于GBRBM-DBN模型的短时交通流预测方法[J].交通信息与安全,2018,36(5):99-108.FENG Wei,CHEN Hong,ZHANG Zhaojin,et al.A forecast of short-term traffic flow based on gbrbm-dbn model [J].Journal of Transport Information and Safety,2018,36(5):99-108.(in Chinese)
[7] 李欣,罗庆,孟德友.基于时空权重相关性的交通流大数据预测方法[J].北京大学学报(自然科学版),2017,53(4):775-782.LI Xin,LUO Qing,MENG Deyou.Traffic flow-big data forecasting method based on spatial-temporal weight correlation [J].Journal of Peking University (Natural Science Edition),2017,53(4):775-782.(in Chinese)
[8] 伍元凯.基于动态张量填充的短时交通流预测研究[D].北京:北京理工大学,2015.WU Yuankai.Short-term traffic prediction based on dynamic tensor completion [D].Beijing:Beijing Institute of Technology,2015.(in Chinese)
[9] 段金肖,丁川,鹿应荣,等.考虑动态波动性的轨道交通站点短时客流预测方法[J].交通信息与安全,2017,35(5):62-69.DUAN Jinxiao,DING chuan,LU Yingrong,et al.A prediction approach of short-term passenger flow of rail transit considering dynamic volatility [J].Journal of Transport Information and Safety,2017,35(5):62-69.(in Chinese)
[10] 李国锋,方正鹏,龚柏岩.带云台摄像机异常交通事件视频识别技术研究[J].公路交通科技(应用技术版),2007(11):168-169,173.LI Guofeng,FANG Zhengpeng,GONG Baiyan.Research on video recognition technology of abnormal traffic incidents with cloud platform camera [J].Highway Traffic Science and Technology (Applied Technology Edition),2007(11):168-169,173.(in Chinese)
[11] 帅丹.道路交通异常事件检测及关键帧提取算法研究[D].重庆:重庆交通大学,2016.SHUAI Dan.The research about the abnormal events detection and the key frame extraction algorithm for traffic road[D].Chongqing:Chongqing Jiaotong University,2016.(in Chinese)
[12] 韩旭.基于车辆轨迹多特征的聚类分析及异常检测方法的研究[D].哈尔滨:哈尔滨工程大学,2014.HAN Xu.Research on cluster analysis and anomaly detection based on multi-feature vehicle trajectory [D].Harbin:Harbin Engineering University,2014.(in Chinese)
[13] 叶敏.基于轨迹数据挖掘的异常检测方法研究[D].西安:长安大学,2017.YE Min.Research on anomaly detection method based on trajectory data mining [D].Xi′an:Chang′an University,2017.(in Chinese)
[14] 王英会.高速公路交通流异常数据识别及修复方法研究[D].北京:北京交通大学,2015.WANG Yinghui.Research on identification and repair methods for abnormal highway traffic flow data [D].Beijing:Beijing Jiaotong University,2015.(in Chinese)
[15] 李明之,马志强,单勇,等.基于轨迹分析的交通目标异常行为识别[J].电视技术,2012,36(1):106-112.LI Mingzhi,MA Zhiqiang,SHAN Yong,et al.Traffic object′s abnormal behavior recognition based on trajectory analysis [J].Video Engineering,2012,36(1):106-112.(in Chinese)
[16] 刘炜华.集群行为异常检测及交通流预测[D].上海:复旦大学,2012.LIU Weihua.Cluster behavior abnormality detection and traffic flow forecasting [D].Shanghai:Fudan University,2012.(in Chinese)
[17] 罗例东.高速公路异常事件影响范围演化分析与预测研究[D].重庆:重庆大学,2015.LUO Zhandong.Study on evolution analysis and prediction of abnormal events influence scope on expressway[D].Chongqing:Chongqing University,2015.(in Chinese)
[18] 王雷,安实,杨海强,等.基于出租车轨迹数据的交通异常识别算法[J].科学技术与工程,2018,18(32):239-247.WANG Lei,AN Shi,YANG Haiqiang,et al.Traffic anomaly recognition algorithm based on taxi track date [J].Science and Technology and Engineering,2018,18(32):239-247.(in Chinese)
[2] 马玄.城市道路交通流预测与状态判别关键技术研究[D].广州:华南理工大学,2014.MA Xuan.Research on the key technologies of urban road traffic flow prediction and state discrimination[D].Guangzhou:South China University of Technology,2014.(in Chinese)
[3] 谷远利,张源,芮小平.基于免疫算法优化LSSVM的短时交通流预测研究[J/OL].(2018-09-18)[2018-11-16].https://doi.org/10.13229/j.cnki.jdxbgxb20180540.GU Yuanli,ZHANG Yuan,RUI Xiaoping.A short-term traffic flow prediction based on lssvm optimized by immune algorithm [J/OL].(2018-09-18)[2018-11-16].https://doi.org/10.13229/j.cnki.jdxbgxb 20180540.(in Chinese)
[4] 王体迎,时鹏超,刘蒋琼,等.基于门限递归单元循环神经网络的交通流预测方法研究[J].重庆交通大学学报(自然科学版),2018,37(11):76-82.WANG Tiying,SHI Pengchao,LIU Jiangqiong,et al.Research on traffic flow prediction method based on gated recurrent unit recurrent neural network [J].Journal of Chongqing Jiaotong University (Natural Science Edition),2018,37(11)76-62.(in Chinese)
[5] 韩春颖,周亚同,常和玲,等.基于稀疏高斯过程混合模型的短时交通流预测[J].交通信息与安全,2019,37(1):121-127.HAN Chunying,ZHOU Yatong,CHANG He Ling,CHI Yue,HE Jingfei.A short-term forecasting of traffic flow based on a mixture model of sparse gaussian process [J].Journal of Transport Safety,2019,37(1):121-127.(in Chinese)
[6] 冯微,陈红,张兆津,等.基于GBRBM-DBN模型的短时交通流预测方法[J].交通信息与安全,2018,36(5):99-108.FENG Wei,CHEN Hong,ZHANG Zhaojin,et al.A forecast of short-term traffic flow based on gbrbm-dbn model [J].Journal of Transport Information and Safety,2018,36(5):99-108.(in Chinese)
[7] 李欣,罗庆,孟德友.基于时空权重相关性的交通流大数据预测方法[J].北京大学学报(自然科学版),2017,53(4):775-782.LI Xin,LUO Qing,MENG Deyou.Traffic flow-big data forecasting method based on spatial-temporal weight correlation [J].Journal of Peking University (Natural Science Edition),2017,53(4):775-782.(in Chinese)
[8] 伍元凯.基于动态张量填充的短时交通流预测研究[D].北京:北京理工大学,2015.WU Yuankai.Short-term traffic prediction based on dynamic tensor completion [D].Beijing:Beijing Institute of Technology,2015.(in Chinese)
[9] 段金肖,丁川,鹿应荣,等.考虑动态波动性的轨道交通站点短时客流预测方法[J].交通信息与安全,2017,35(5):62-69.DUAN Jinxiao,DING chuan,LU Yingrong,et al.A prediction approach of short-term passenger flow of rail transit considering dynamic volatility [J].Journal of Transport Information and Safety,2017,35(5):62-69.(in Chinese)
[10] 李国锋,方正鹏,龚柏岩.带云台摄像机异常交通事件视频识别技术研究[J].公路交通科技(应用技术版),2007(11):168-169,173.LI Guofeng,FANG Zhengpeng,GONG Baiyan.Research on video recognition technology of abnormal traffic incidents with cloud platform camera [J].Highway Traffic Science and Technology (Applied Technology Edition),2007(11):168-169,173.(in Chinese)
[11] 帅丹.道路交通异常事件检测及关键帧提取算法研究[D].重庆:重庆交通大学,2016.SHUAI Dan.The research about the abnormal events detection and the key frame extraction algorithm for traffic road[D].Chongqing:Chongqing Jiaotong University,2016.(in Chinese)
[12] 韩旭.基于车辆轨迹多特征的聚类分析及异常检测方法的研究[D].哈尔滨:哈尔滨工程大学,2014.HAN Xu.Research on cluster analysis and anomaly detection based on multi-feature vehicle trajectory [D].Harbin:Harbin Engineering University,2014.(in Chinese)
[13] 叶敏.基于轨迹数据挖掘的异常检测方法研究[D].西安:长安大学,2017.YE Min.Research on anomaly detection method based on trajectory data mining [D].Xi′an:Chang′an University,2017.(in Chinese)
[14] 王英会.高速公路交通流异常数据识别及修复方法研究[D].北京:北京交通大学,2015.WANG Yinghui.Research on identification and repair methods for abnormal highway traffic flow data [D].Beijing:Beijing Jiaotong University,2015.(in Chinese)
[15] 李明之,马志强,单勇,等.基于轨迹分析的交通目标异常行为识别[J].电视技术,2012,36(1):106-112.LI Mingzhi,MA Zhiqiang,SHAN Yong,et al.Traffic object′s abnormal behavior recognition based on trajectory analysis [J].Video Engineering,2012,36(1):106-112.(in Chinese)
[16] 刘炜华.集群行为异常检测及交通流预测[D].上海:复旦大学,2012.LIU Weihua.Cluster behavior abnormality detection and traffic flow forecasting [D].Shanghai:Fudan University,2012.(in Chinese)
[17] 罗例东.高速公路异常事件影响范围演化分析与预测研究[D].重庆:重庆大学,2015.LUO Zhandong.Study on evolution analysis and prediction of abnormal events influence scope on expressway[D].Chongqing:Chongqing University,2015.(in Chinese)
[18] 王雷,安实,杨海强,等.基于出租车轨迹数据的交通异常识别算法[J].科学技术与工程,2018,18(32):239-247.WANG Lei,AN Shi,YANG Haiqiang,et al.Traffic anomaly recognition algorithm based on taxi track date [J].Science and Technology and Engineering,2018,18(32):239-247.(in Chinese)
© 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 |