Multi-population genetic neural network model for short-term traffic flow prediction at intersections
文献类型：EI期刊论文
| 作者 | Zhu Axing |
| 发表日期 | 2009 |
| 关键词 | Backpropagation algorithms Genetic algorithms Global optimization Intelligent vehicle highway systems Intersections Linearization Neural networks Traffic surveys Vehicle locating systems |
| 英文摘要 | As the basis and a key issue of intelligent transportation system (ITS), researches on approaches for traffic flow prediction are significant. To adapt to the real-time characteristics and nonlinear feature of short-term traffic flow prediction at intersections, this paper presents a Multi-Population Genetic Neural Network (MPGNN) model for traffic flow prediction. MPGNN model combines the ability of BP neural network for solving nonlinear problem and the ability of genetic algorithm (GA) for global optimization. And in this model, multiple populations of GA are built to search an optimized structure of BP network. A case study using this model is conducted at the intersection of Luoyu Road, Wuluo Road, South Luoshi Road and North Luoshi Road. This study shows that the predicted short-term traffic flow is accurate and the MPGNN model is effective. Based on the predicted results to control and reassign the traffic flow on these roads, the traffic pressure of these roads in rush hours can be greatly alleviated. |
| 出处 | Cehui Xuebao/Acta Geodaetica et Cartographica Sinica
 |
| 卷 | 38期:4页:363-368 |
| 收录类别 | EI |
| 语种 | 英语 |
| 源URL | [http://ir.igsnrr.ac.cn/handle/311030/24757]  |
| 专题 | 地理科学与资源研究所_历年回溯文献 |
| 推荐引用方式 GB/T 7714 | Zhu Axing. Multi-population genetic neural network model for short-term traffic flow prediction at intersections. 2009. |
入库方式： OAI收割
来源：地理科学与资源研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。