| 经营性高速公路停止收费后运营成本预测模型 |
| |
| 引用本文： | 章权,周伟,温惠英,林珊.经营性高速公路停止收费后运营成本预测模型[J].长安大学学报(自然科学版),2011(6):90-93,104. |
| |
| 作者姓名： | 章权 周伟 温惠英 林珊 |
| |
| 作者单位： | 长安大学公路学院;广东省公路管理局;华南理工大学土木与交通学院; |
| |
| 基金项目： | 国家自然科学基金项目(50978106) |
| |
| 摘 要： | 针对单纯的神经网络会陷入局部最小点的缺点,以及单纯的GA方法不具备自适应学习能力,且在处理局部区域上存在一定问题的缺点,通过将遗传算法与神经网络相结合,提出了遗传算法优化神经网络权重的混合算法,并将此模型用于对经营性高速公路收费期满、停止收费后的运营成本进行预测。研究结果表明:相比单纯的BP算法,该算法能同时对解空间内的多个点进行遗传优选,避免陷入局部最小点,具有更快的收敛速度和更高的预测精度;该算法预测平均误差为0.06%,比单纯的BP算法提高了1.5%;利用该算法进行经营性高速公路停止收费后运营成本预测,可为高速公路运营管理者进行科学决策提供参考。
|
| 关 键 词： | 交通工程 高速公路 运营成本 遗传算法 神经网络 预测 |
Operation cost prediction model of expressway after canceling toll |
| |
| ZHANG Quan,ZHOU Wei,WEN Hui-ying,LIN Shan.Operation cost prediction model of expressway after canceling toll[J].JOurnal of Chang’an University:Natural Science Edition,2011(6):90-93,104. |
| |
| Authors: | ZHANG Quan ZHOU Wei WEN Hui-ying LIN Shan |
| |
| Institution: | ZHANG Quan1,2,ZHOU Wei1,WEN Hui-ying3,LIN Shan3(1.School of Highway,Chang'an University,Xi'an 710064,Shaanxi,China,2.Guangdong Province Administration of Highway,Guangzhou 510075,Guangdong,3.School of Civil Engineering and Transportation,South China University of Technology,Guangzhou 510640,China) |
| |
| Abstract: | In order to avoid the shortcomings of simple neural network that will fall into local minimum point,through combing genetic algorithm with neural network,this paper proposed a hybrid algorithm which using genetic algorithm to optimize the weight of neural network,and used it to predict operating costs of expressway after canceling toll.The results show that the method can predict the operating costs better and also has higher prediction accuracy than simple BP neural network method,the average forecast erro... |
| |
| Keywords: | traffic engineering expressway operation cost genetic algorithm neural network prediction |
| 本文献已被 CNKI 等数据库收录！ |
|
