基于混沌算法的交通流量车辆调度过程仿真
发布时间：2018-09-10 20:12
【摘要】：研究交通车辆的高效调度的问题。随着车辆交通复杂程度的增加,调度过程中车辆的可调度流量特征变得复杂,呈现了非线性变化。传统的车辆调度方法都是以流量特征为基础进行调节,受到交通流量的非线性变化影响,效率偏低。为解决上述问题,在传统的小波神经网络预测交通调度模型中,引入了混沌重构的混沌时间序列预测,并且加入一种基于混沌算法的快速交通信息学习算法,在实际小波神经网络调度模型的建立中,以交通流量混沌时间序列为基础,确定网络模型的输入神经元个数、隐含层个数以及神经元个数,克服非线性的干扰。仿真结果表明,改进模型对车辆的调度精度和改进效果较好。
[Abstract]:The problem of efficient dispatching of traffic vehicles is studied. With the increase of vehicle traffic complexity, the characteristics of vehicle schedulable flow become complex and nonlinear. The traditional vehicle scheduling methods are based on the characteristics of traffic flow, which is influenced by the nonlinear change of traffic flow, and the efficiency is low. In order to solve the above problems, chaotic time series prediction based on chaotic reconstruction is introduced into the traditional wavelet neural network traffic scheduling model, and a fast traffic information learning algorithm based on chaotic algorithm is added. Based on the chaotic time series of traffic flow, the number of input neurons, the number of hidden layers and the number of neurons in the network model are determined to overcome the nonlinear interference in the establishment of the actual wavelet neural network scheduling model. The simulation results show that the improved model has better precision and better effect on vehicle scheduling.
【作者单位】： 广东外语外贸大学南国商学院信息科学技术系;华南理工大学计算机科学与工程学院;
【基金】：国家自然科学基金资助项目(61171141) 广东省产学研省部合作专项资金项目(2012B091100448)
【分类号】：U492.22
本文编号：2235492
[Abstract]:The problem of efficient dispatching of traffic vehicles is studied. With the increase of vehicle traffic complexity, the characteristics of vehicle schedulable flow become complex and nonlinear. The traditional vehicle scheduling methods are based on the characteristics of traffic flow, which is influenced by the nonlinear change of traffic flow, and the efficiency is low. In order to solve the above problems, chaotic time series prediction based on chaotic reconstruction is introduced into the traditional wavelet neural network traffic scheduling model, and a fast traffic information learning algorithm based on chaotic algorithm is added. Based on the chaotic time series of traffic flow, the number of input neurons, the number of hidden layers and the number of neurons in the network model are determined to overcome the nonlinear interference in the establishment of the actual wavelet neural network scheduling model. The simulation results show that the improved model has better precision and better effect on vehicle scheduling.
【作者单位】： 广东外语外贸大学南国商学院信息科学技术系;华南理工大学计算机科学与工程学院;
【基金】：国家自然科学基金资助项目(61171141) 广东省产学研省部合作专项资金项目(2012B091100448)
【分类号】：U492.22
【相似文献】
相关期刊论文 前3条
1 苏绍娟;王丽铮;王呈方;;不确定性航线配船数学模型建模方法[J];船海工程;2007年04期
2 董超俊;刘智勇;刘贤坤;;基于粗糙集的区域交通控制交通量属性约简[J];系统仿真学报;2006年06期
3 宋运林;董超俊;;混沌蚁群算法及其在相位差优化中应用[J];科技信息(科学教研);2007年35期
相关硕士学位论文 前1条
1 宋运林;城市交通控制信号配时技术研究[D];五邑大学;2008年
，本文编号：2235492
本文链接：https://www.wllwen.com/kejilunwen/jiaotonggongchenglunwen/2235492.html
