基于CNN-SVR混合深度学习模型的短时交通流预测
|
摘要
精准且快速的短时交通流预测是智能交通发展的重要组成部分.本文针对当前交通流预测模型不能充分提取交通流数据的时空特征、预测性能容易受到外界干扰因素影响的问题,提出一种基于深度学习的短时交通流预测模型,该模型结合卷积神经网络(Convolutional Neural Network,CNN)与支持向量回归分类器(Support Vector Regression,SVR)的特点:在网络底层应用CNN进行交通流特征提取,并将提取结果输入到SVR回归模型中进行流量预测.为验证模型的有效性,取G103国道的实际交通流量数据进行试验.结果表明,提出的预测模型与传统的预测模型相比具有更高的预测精度,预测性能提高了11%,是一种有效的交通流预测模型.
It is very important for intelligent transportation development to realize accurate and fast traffic forecast. However, dominant models for short-term traffic flow forecasting can't extract spatial-temporal characteristics of traffic flow data amply. Moreover, these models are susceptible to outside factors. To resolve these problems, an innovative model based on deep learning is proposed in this paper. Convolutional Neural Network(CNN) and Support Vector Regression(SVR) classifier are combined in this model: feature learning of traffic flow is carried out by using CNN in underlying network, then the extracted results are transmitted to SVR model as input to predict traffic flow. To verify the validity of the proposed model,experiments are conducted on actual traffic flow data of China national highway 103(G103). Experimental results show that the proposed model has higher prediction accuracy than the traditional prediction model,and the prediction performance is improved by 11%, which is an effective traffic flow forecasting model.
It is very important for intelligent transportation development to realize accurate and fast traffic forecast. However, dominant models for short-term traffic flow forecasting can't extract spatial-temporal characteristics of traffic flow data amply. Moreover, these models are susceptible to outside factors. To resolve these problems, an innovative model based on deep learning is proposed in this paper. Convolutional Neural Network(CNN) and Support Vector Regression(SVR) classifier are combined in this model: feature learning of traffic flow is carried out by using CNN in underlying network, then the extracted results are transmitted to SVR model as input to predict traffic flow. To verify the validity of the proposed model,experiments are conducted on actual traffic flow data of China national highway 103(G103). Experimental results show that the proposed model has higher prediction accuracy than the traditional prediction model,and the prediction performance is improved by 11%, which is an effective traffic flow forecasting model.
引文
[1]HUANG W,SONG G,HONG H,et al.Deep architecture for traffic flow prediction:Deep belief networks with multitask learning[J].IEEE Transactions on Intelligent Transportation Systems,2014,15(5):2191-2201.
[2]LV Y,DUAN Y,KANG W,et al.Traffic flow prediction with big data:A deep learning approach[J].IEEETransactions on Intelligent Transportation Systems,2015,16(2):865-873.
[3]YANG H F,DILLON T S,CHEN Y P.Optimized structure of the traffic flow forecasting model with a deep learning approach[J].IEEE Transactions on Neural Networks&Learning Systems.in press,doi:10.1109/TNNLS.2016.2574840.
[4]WU Y,TAN H.Short-term traffic flow forecasting with spatial-temporal correlation in a hybrid deep learning framework[J/OL].(2016-12-3)[2017-3-10].https://arxiv.org/abs/1612.01022.
[5]JIA Y,WU J,DU Y.Traffic speed prediction using deep learning method[C]//2016 IEEE 19th International Conference on Intelligent Transportation Systems(ITSC).Piscataway,NJ,USA:IEEE,2016:1217-1222.
[6]FUSCO G,COLOMBARONI C,ISAENKO C.Shortterm speed predictions exploiting big data on large urban road networks[J].Transportation Research,Part C:Emerging Technologies,2016(73):183-201.
[7]WANG J Y,GU Q,WU J J,et al.Traffic speed prediction and congestion source exploration:A Deep Learning Method[C]//Proceedings-IEEE International Conference on Data Mining,ICDM.Barcelona,Spain:Institute of Electrical and Electronics Engineers Inc.,2017:499-508.
[8]LOLLI F,GUMBERINI R,REGATTIERI A,et al.Single-hidden layer neural networks for forecasting intermittent demand[J].Production Economics,2017(183):116-128.
[9]TANG Y.Deep learning using linear support vector machines[J/OL].(2015-2-21)[2017-3-10].https://arxiv.org/abs/1612.01022.
[10]LI W,HU J,CHEN B H.A deep quasi-linear kernel composition method for support vector machines[C]//Proceedings of the International Joint Conference on Neural Networks.Vancouver,Canada:Institute of Electrical and Electronics Engineers Inc.,2016:1639-1645.
[11]傅贵,韩国强,逯峰,等.基于支持向量机回归的的短时交通流预测模型[J].华南理工大学学报(自然科学版).2013,41(9):71-76.[FU G,HAN G Q,LU F,et al.Short-term traffic flow prediction based on support vector machine[J].Journal of South China University of Technoligy(Natural Science Edition),2013,41(9):71-76.]
[12]曾绍华.支持向量回归机算法理论研究与应用[D].重庆:重庆大学,2006.[ZENG S H.The theory research of algorithm on support vector regression and application[D].Chongqing:Chongqing University,2006.]
[13]HUANG H,HUANG X,LI R,et al.Sound quality prediction of vehicle noise using deep belief networks[J].Applied Acoustics,2016(113):149-161.
[14]YU D,DENG L.Deep learning and its applications to signal and information processing[exploratory DSP][J].IEEE Signal Processing Magazine.2011,28(1):145-154.
[2]LV Y,DUAN Y,KANG W,et al.Traffic flow prediction with big data:A deep learning approach[J].IEEETransactions on Intelligent Transportation Systems,2015,16(2):865-873.
[3]YANG H F,DILLON T S,CHEN Y P.Optimized structure of the traffic flow forecasting model with a deep learning approach[J].IEEE Transactions on Neural Networks&Learning Systems.in press,doi:10.1109/TNNLS.2016.2574840.
[4]WU Y,TAN H.Short-term traffic flow forecasting with spatial-temporal correlation in a hybrid deep learning framework[J/OL].(2016-12-3)[2017-3-10].https://arxiv.org/abs/1612.01022.
[5]JIA Y,WU J,DU Y.Traffic speed prediction using deep learning method[C]//2016 IEEE 19th International Conference on Intelligent Transportation Systems(ITSC).Piscataway,NJ,USA:IEEE,2016:1217-1222.
[6]FUSCO G,COLOMBARONI C,ISAENKO C.Shortterm speed predictions exploiting big data on large urban road networks[J].Transportation Research,Part C:Emerging Technologies,2016(73):183-201.
[7]WANG J Y,GU Q,WU J J,et al.Traffic speed prediction and congestion source exploration:A Deep Learning Method[C]//Proceedings-IEEE International Conference on Data Mining,ICDM.Barcelona,Spain:Institute of Electrical and Electronics Engineers Inc.,2017:499-508.
[8]LOLLI F,GUMBERINI R,REGATTIERI A,et al.Single-hidden layer neural networks for forecasting intermittent demand[J].Production Economics,2017(183):116-128.
[9]TANG Y.Deep learning using linear support vector machines[J/OL].(2015-2-21)[2017-3-10].https://arxiv.org/abs/1612.01022.
[10]LI W,HU J,CHEN B H.A deep quasi-linear kernel composition method for support vector machines[C]//Proceedings of the International Joint Conference on Neural Networks.Vancouver,Canada:Institute of Electrical and Electronics Engineers Inc.,2016:1639-1645.
[11]傅贵,韩国强,逯峰,等.基于支持向量机回归的的短时交通流预测模型[J].华南理工大学学报(自然科学版).2013,41(9):71-76.[FU G,HAN G Q,LU F,et al.Short-term traffic flow prediction based on support vector machine[J].Journal of South China University of Technoligy(Natural Science Edition),2013,41(9):71-76.]
[12]曾绍华.支持向量回归机算法理论研究与应用[D].重庆:重庆大学,2006.[ZENG S H.The theory research of algorithm on support vector regression and application[D].Chongqing:Chongqing University,2006.]
[13]HUANG H,HUANG X,LI R,et al.Sound quality prediction of vehicle noise using deep belief networks[J].Applied Acoustics,2016(113):149-161.
[14]YU D,DENG L.Deep learning and its applications to signal and information processing[exploratory DSP][J].IEEE Signal Processing Magazine.2011,28(1):145-154.