基于Petri网建模的道路车流量预测方法研究
|
摘要
在对基于Petri网建模道路进行分析时发现,当利用智能算法优化交通信号时会存在一定滞后性,主要原因在于智能算法运算效率不高,以及道路车辆产生的一些随机因素等。因此,使用Petri网对交叉口进行建模并模拟车流产生,并运用改进卡尔曼滤波算法对道路车流量进行预测,用于弥补因车辆随机因素或使用智能算法造成的时滞。改进算法采用SVM回归对利用卡尔曼滤波产生的误差进行拟合与预测,用于补偿卡尔曼滤波算法产生的误差。实验结果表明,利用SVM滤波预测误差降低了6个百分点,证明该方法有效提高了车流预测精度。
When analyzing the road modeled by Petri net,we find that there are always some lags when using intelligent algorithm to optimize traffic signals. The main reason is the efficiency of intelligent algorithm or some random factors of the road,etc. There fore this paper use the improved Kalman filter algorithm to predict the traffic flow rate of road based on Petri net. It is used to compensate the time delay caused by intelligent algorithm or random factors. Then we use the SVM regression method to estimate and compensate the prediction error of Kalman filter. The experiment shows that the total error of the prediction is 6 percentage points higher than that of the conventional Kalman filtering. It is proved that this method can improve the accuracy of traffic flow prediction.
When analyzing the road modeled by Petri net,we find that there are always some lags when using intelligent algorithm to optimize traffic signals. The main reason is the efficiency of intelligent algorithm or some random factors of the road,etc. There fore this paper use the improved Kalman filter algorithm to predict the traffic flow rate of road based on Petri net. It is used to compensate the time delay caused by intelligent algorithm or random factors. Then we use the SVM regression method to estimate and compensate the prediction error of Kalman filter. The experiment shows that the total error of the prediction is 6 percentage points higher than that of the conventional Kalman filtering. It is proved that this method can improve the accuracy of traffic flow prediction.
引文
[1]石曼曼.基于卡尔曼滤波的短时交通流预测方法研究[D].成都:西南交通大学,2012.
[2]黄辉先,史忠科.城市单交叉路口交通流实时遗传算法优化控制[J].系统工程理论与实践,2001(3):102-106.
[3]FEBBRARO A D,GIGLIO D,SACCO N.A deterministic and stochastic petri Net model for traffic-responsive signaling control in urban areas[J].IEEE Transactions on Intelligent Transportation Systems,2016,17(2):510-524.
[4]JULVEZ J,BOEL R K.A continuous petri net approach for model predictive control of traffic systems[J].Transactions on System Man and Cybernetics Part A-systems and Humans,2010,40(4):686-697.
[5]WANG J,YAN J,LI L.Microscopic modeling of a signalized traffic intersection using timed Petri nets[J].IEEE Transactions on Intelligent Transportation Systems,2016,17(2):305-312.
[6]郑子茹.车路协同系统中无信号交叉口优化控制方法研究[D].北京:北京交通大学,2015.
[7]杨兆升,朱中.基于卡尔曼滤波理论的交通流量实时预测模型[J].中国公路学报,1999,12(3):63-67.
[8]郭海锋,方良君,俞立.基于模糊卡尔曼滤波的短时交通流量预测方法[J].浙江工业大学学报,2013(2):218-221.
[9]傅贵,韩国强,逯峰.基于支持向量机回归的短时交通流预测模型[J].华南理工大学学报,2013(9):71-76.
[10]赵卓峰,杨宗润.基于残差修正GM(1,1)模型的车流量预测[J].计算机科学,2017(4):96-99.
[11]KIM W,PARK B H.Multi-time step traffic queue prediction model based on discrete time point process(DTPP)[C].International Conference on Future Generation Communication&Networking Symposia.IEEE,2008.
[12]孙冰.城市道路交通单向拥堵点预测分流仿真[J].计算机仿真,2017(12):139-142.
[13]吴哲辉.Petri网导论[M].北京:机械工业出版社,2006.
[14]宋文尧,张牙.卡尔曼滤波[M].北京:科学出版社,1991.
[15]许珊.基于交通流预测的单交叉口信号控制研究[D].大庆:东北石油大学,2014.
[16]林冬霞.基于车联网的短时交通流预测算法研究[D].广州:华南理工大学,2015.
[17]王凡.基于支持向量机的交通流预测方法研究[D].大连:大连理工大学,2010.
[18]李斌,郗涛,史明华.基于支持向量机的交通流组合预测模型[J].天津工业大学学报,2008,27(2):73-77.
[19]杨兆升,王媛,管青.基于支持向量机方法的短时交通流量预测方法[J].吉林大学学报,2006(6):881-884.
[20]ANDREW NG.最优间隔分类器问题[EB/OL].http://open.163.com/special/opencourse/machinelearning.html.
[2]黄辉先,史忠科.城市单交叉路口交通流实时遗传算法优化控制[J].系统工程理论与实践,2001(3):102-106.
[3]FEBBRARO A D,GIGLIO D,SACCO N.A deterministic and stochastic petri Net model for traffic-responsive signaling control in urban areas[J].IEEE Transactions on Intelligent Transportation Systems,2016,17(2):510-524.
[4]JULVEZ J,BOEL R K.A continuous petri net approach for model predictive control of traffic systems[J].Transactions on System Man and Cybernetics Part A-systems and Humans,2010,40(4):686-697.
[5]WANG J,YAN J,LI L.Microscopic modeling of a signalized traffic intersection using timed Petri nets[J].IEEE Transactions on Intelligent Transportation Systems,2016,17(2):305-312.
[6]郑子茹.车路协同系统中无信号交叉口优化控制方法研究[D].北京:北京交通大学,2015.
[7]杨兆升,朱中.基于卡尔曼滤波理论的交通流量实时预测模型[J].中国公路学报,1999,12(3):63-67.
[8]郭海锋,方良君,俞立.基于模糊卡尔曼滤波的短时交通流量预测方法[J].浙江工业大学学报,2013(2):218-221.
[9]傅贵,韩国强,逯峰.基于支持向量机回归的短时交通流预测模型[J].华南理工大学学报,2013(9):71-76.
[10]赵卓峰,杨宗润.基于残差修正GM(1,1)模型的车流量预测[J].计算机科学,2017(4):96-99.
[11]KIM W,PARK B H.Multi-time step traffic queue prediction model based on discrete time point process(DTPP)[C].International Conference on Future Generation Communication&Networking Symposia.IEEE,2008.
[12]孙冰.城市道路交通单向拥堵点预测分流仿真[J].计算机仿真,2017(12):139-142.
[13]吴哲辉.Petri网导论[M].北京:机械工业出版社,2006.
[14]宋文尧,张牙.卡尔曼滤波[M].北京:科学出版社,1991.
[15]许珊.基于交通流预测的单交叉口信号控制研究[D].大庆:东北石油大学,2014.
[16]林冬霞.基于车联网的短时交通流预测算法研究[D].广州:华南理工大学,2015.
[17]王凡.基于支持向量机的交通流预测方法研究[D].大连:大连理工大学,2010.
[18]李斌,郗涛,史明华.基于支持向量机的交通流组合预测模型[J].天津工业大学学报,2008,27(2):73-77.
[19]杨兆升,王媛,管青.基于支持向量机方法的短时交通流量预测方法[J].吉林大学学报,2006(6):881-884.
[20]ANDREW NG.最优间隔分类器问题[EB/OL].http://open.163.com/special/opencourse/machinelearning.html.