基于变分BP神经网络的普通公路轴载计算  

Axle Load Calculation of Ordinary Highway Based on Variational BP Neural Network

在线阅读下载全文

作  者:刘坤 张路凯 孙满 撒蕾[2] 王英平[2] 张金萌 LIU Kun;ZHANG Lu-kai;SUN Man;SA Lei;WANG Ying-ping;ZHANG Jin-meng(Xuzhou Highway Development Center,Xuzhou,Jiangsu 221006,China;Transport Planning and Research Institute,Ministry of Transport,Beijing 100029,China)

机构地区:[1]徐州市公路事业发展中心,江苏徐州221006 [2]交通运输部规划研究院,北京100029

出  处:《黑龙江交通科技》2023年第11期72-77,81,共7页Communications Science and Technology Heilongjiang

基  金:江苏省交通运输科技项目“基于轴载调查的普通公路重载交通特征分析研究”;交通运输部规划研究院科技开发项目“公路交通轴载估计模型”

摘  要:以普通公路为研究对象,旨在实现路段轴载谱和当量轴次的计算。首先,以《公路沥青路面设计规范》为依据,分析轴载谱和累计当量轴次的计算需求;之后,针对行业信息资源现状,解析多输入多输出的数据驱动特征;进而,针对多元耦合的计算流程,建立变分BP神经网络模型;最后,以徐州地区普通公路网为实例,对所提出方法进行验证。验证集计算中,所得月度当量轴次的推算偏差为17.15%,优于灰色系统预测模型偏差结果(19.89%)以及多元线性回归模型偏差结果(22.37%);轴载谱推算结果中,精度最高的车型为3轴车。所计算的重点轴荷范围可作为重载交通管控和基建养护决策的参考。Taking ordinary roads as the object,this study aims to realize the calculation of the axle load spectrum and equivalent axle times of the road section.Firstly,based on the Code for Design of Highway Asphalt Pavement,the calculation requirements of axle load spectrum and cumulative equivalent axle times were analyzed.Then,according to the current situation of industry information resources,the data-driven characteristics of multiple inputs and multiple outputs were analyzed.The calculation process was used to build a variational BP neural network model.Finally,the proposed method was verified by taking the ordinary highway network in Xuzhou area as an example.In the calculation of the validation set,the calculated deviations of the obtained monthly equivalent axis times were 17.15%,which were better than the deviation results of the gray system prediction model(19.89%)and the deviation results of the multiple linear regression model(22.37%).In the calculation results of axle load spectrum,the model with the highest accuracy was the 3-axle vehicle.The calculated range of key axle loads can be used as a reference for heavy-load traffic control and infrastructure maintenance decisions.

关 键 词:重载交通 普通公路 变分BP神经网络 轴载谱 当量轴次 

分 类 号:U416.02[交通运输工程—道路与铁道工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象