基于IPSO-BP神经网络的短时交通流量预测
|
摘要
短时交通流量具有非线性、随机性等特点,如何准确地进行短时交通流量预测,是智能交通系统研究的一项关键内容。传统的预测模型不能实时反映短时交通流量变化特点,同时BP神经网络的交通流量预测存在收敛速度缓慢、易陷入局部极值、预测精度低等缺点。为了提高短时交通流量预测精度,提出了一种基于改进粒子群算法(IPSO)优化BP神经网络的复合预测模型,引入相对误差指标作为预测模型的评价指标,并利用实测的道路短时交通流数据对所构建的预测模型进行验证。结果表明,所提出的预测模型在短时间内寻出全局最优解,具有较好的预测精度,提高了短时交通流量预测的准确性和可靠性。
The short-time traffic flow has the characteristics of nonlinearity and randomness. How to predict the shortterm traffic flow accurately is a key content in the research of intelligent transportation system. The traditional prediction model can not reflect the characteristics of short-term traffic flow in real time. Meanwhile,the BP neural network has some shortcomings such as slow convergence rate,easy to fall into local extreme value and low prediction accuracy. In order to improve the accuracy of short-term traffic flow prediction,a composite prediction model based on improves particle swarm potimization( IPSO) algorithm for BP netural network is proposed. The relative error index is introduced as the evaluation index of the prediction model,and the measured short-term traffic flow data is used. The prediction model is validated. The results show that the proposed prediction model searches for the global optimal solution in a short time,which has better prediction accuracy and improves the accuracy and reliability of short-term traffic flow prediction.
The short-time traffic flow has the characteristics of nonlinearity and randomness. How to predict the shortterm traffic flow accurately is a key content in the research of intelligent transportation system. The traditional prediction model can not reflect the characteristics of short-term traffic flow in real time. Meanwhile,the BP neural network has some shortcomings such as slow convergence rate,easy to fall into local extreme value and low prediction accuracy. In order to improve the accuracy of short-term traffic flow prediction,a composite prediction model based on improves particle swarm potimization( IPSO) algorithm for BP netural network is proposed. The relative error index is introduced as the evaluation index of the prediction model,and the measured short-term traffic flow data is used. The prediction model is validated. The results show that the proposed prediction model searches for the global optimal solution in a short time,which has better prediction accuracy and improves the accuracy and reliability of short-term traffic flow prediction.
引文
[1]刘静,李亮,关伟,等.基于神经网络的北京环路交通流短期预测研究[J].交通运输系统工程与信息.2005,5(6):110-115.
[2]张振洪.道路交通流短时预测算法研究及系统应用实现[D].广州:华南理工大学,2015.
[3]董春娇,邵春福,周雪梅,等.基于交通流参数相关的阻塞流短时预测卡尔曼滤波算法[J].东南大学学报:自然科学版,2014,44(2):413-418.
[4]傅贵,韩国强,逯峰,等.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:自然科学版,2013,41(9):71-76.
[5]聂佩林,余志,何兆成.基于约束卡尔曼滤波的短时交通流量组合预测模型[J].交通运输工程学报,2008,8(5):86-90.
[6]丁恒,郑小燕,刘燕,等.基于小波分析的随机交通流组合预测方法[J].系统仿真学报,2012,24(2):377-381.
[7]杨高飞,徐睿,秦鸣,等.基于ARMA和卡尔曼滤波的短时交通预测[J].郑州大学学报:工学版,2017,38(2):37-40.
[8]卢建中,程浩.改进GA优化BP神经网络的短时交通流预测[J].合肥工业大学学报:自然科学版,2015,38(1):127-131.
[9]张春生,刘树东,谭覃.基于改进遗传算法优化BP神经网络的短时交通流量预测方法研究[J].天津城建大学学报,2017,23(2):143-147.
[10]张军,王远强,朱新山.改进PSO优化神经网络的短时交通流预测[J].计算机工程与应用,2017,53(14):227-231.
[11]田景文,高美娟.人工神经网络算法研究及应用[M].北京:北京理工大学出版社,2006.
[12]HE K,ZHANG X,REN S,et al.Delving deep into rectifiers:surpassing human-level performance on Image Net classification[C]//Proceedings of the2015 IEEE International Conference on Computer Vision,Santiago,Chile,December 7-13,2015:1026-1034.
[13]DONAHUE J,JIA Y,VINYALS O,et al.DeCAF:a deep convolution activation feature for generic visual recognition[J].Computer Science,2013,50(1):815-830.
[14]马军杰,尤建新,陈震.基于改进粒子群优化算法的灰色神经网络模型[J].同济大学学报,2012,40(5):740-743.
[15]雷斌,陶海龙,徐晓光.基于改进粒子群优化算法的灰色神经网络的铁路货运量预测[J].计算机应用,2012,32(10):2948-2951.
[16]孟飞,兰巨龙,胡宇翔.基于改进的量子粒子群优化小波神经网络的网络流量预测[J].计算机应用研究,2015,32(5):1450-1453.
[17]范庆波,江福才,马全党,等.基于PSO的BP神经网络-Markov船舶交通流量预测模型[J].上海海事大学学报,2018,39(2):22-27.
[18]韦凌翔,陈红,王永岗.基于RVM和ARIMA的短时交通流量预测方法研究[J].武汉理工大学学报:交通科学与工程版,2017,41(2):349-354.
[19]王建,邓卫,赵金宝.基于改进型贝叶斯组合模型的短时交通流量预测[J].东南大学学报:自然科学版,2012,42(1):162-167.
[2]张振洪.道路交通流短时预测算法研究及系统应用实现[D].广州:华南理工大学,2015.
[3]董春娇,邵春福,周雪梅,等.基于交通流参数相关的阻塞流短时预测卡尔曼滤波算法[J].东南大学学报:自然科学版,2014,44(2):413-418.
[4]傅贵,韩国强,逯峰,等.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:自然科学版,2013,41(9):71-76.
[5]聂佩林,余志,何兆成.基于约束卡尔曼滤波的短时交通流量组合预测模型[J].交通运输工程学报,2008,8(5):86-90.
[6]丁恒,郑小燕,刘燕,等.基于小波分析的随机交通流组合预测方法[J].系统仿真学报,2012,24(2):377-381.
[7]杨高飞,徐睿,秦鸣,等.基于ARMA和卡尔曼滤波的短时交通预测[J].郑州大学学报:工学版,2017,38(2):37-40.
[8]卢建中,程浩.改进GA优化BP神经网络的短时交通流预测[J].合肥工业大学学报:自然科学版,2015,38(1):127-131.
[9]张春生,刘树东,谭覃.基于改进遗传算法优化BP神经网络的短时交通流量预测方法研究[J].天津城建大学学报,2017,23(2):143-147.
[10]张军,王远强,朱新山.改进PSO优化神经网络的短时交通流预测[J].计算机工程与应用,2017,53(14):227-231.
[11]田景文,高美娟.人工神经网络算法研究及应用[M].北京:北京理工大学出版社,2006.
[12]HE K,ZHANG X,REN S,et al.Delving deep into rectifiers:surpassing human-level performance on Image Net classification[C]//Proceedings of the2015 IEEE International Conference on Computer Vision,Santiago,Chile,December 7-13,2015:1026-1034.
[13]DONAHUE J,JIA Y,VINYALS O,et al.DeCAF:a deep convolution activation feature for generic visual recognition[J].Computer Science,2013,50(1):815-830.
[14]马军杰,尤建新,陈震.基于改进粒子群优化算法的灰色神经网络模型[J].同济大学学报,2012,40(5):740-743.
[15]雷斌,陶海龙,徐晓光.基于改进粒子群优化算法的灰色神经网络的铁路货运量预测[J].计算机应用,2012,32(10):2948-2951.
[16]孟飞,兰巨龙,胡宇翔.基于改进的量子粒子群优化小波神经网络的网络流量预测[J].计算机应用研究,2015,32(5):1450-1453.
[17]范庆波,江福才,马全党,等.基于PSO的BP神经网络-Markov船舶交通流量预测模型[J].上海海事大学学报,2018,39(2):22-27.
[18]韦凌翔,陈红,王永岗.基于RVM和ARIMA的短时交通流量预测方法研究[J].武汉理工大学学报:交通科学与工程版,2017,41(2):349-354.
[19]王建,邓卫,赵金宝.基于改进型贝叶斯组合模型的短时交通流量预测[J].东南大学学报:自然科学版,2012,42(1):162-167.