基于BP神经网络的广东省路面使用性能评价  

Evaluation of Pavement Performance in Guangdong Province Based on BP Neural Network

作　　者：张凯星 王选仓[1] 赵静[1] 樊琳 李善强 ZHANG Kai-xing;WANG Xuan-cang;ZHAO Jing;FAN Lin;LI Shan-qiang(School of Highway,Chang’an University,Xi’an 710064,China;China Highway Engineering Consultants Corporation,Beijing 100089,China;Guangdong Hualu Transportation Technology Co.Ltd.,Guangzhou 510420,China)

机构地区：[1]长安大学公路学院,西安市710064 [2]中国公路工程咨询集团有限公司,北京市100089 [3]广东华路交通科技有限公司,广州市510420

出　　处：《公路》2020年第10期309-314,共6页Highway

基　　金：广东省交通运输厅科技项目,项目编号科技-2015-02-011。

摘　　要：路面使用性能评价,是我国路面养护的关键环节,对路面养护管理具有重要意义。为了使路面性能评价更加合理,在规范所提出的综合评价指标的基础上,提出了新的评价方法。该方法利用BP神经网络,同时又考虑了路面结构强度指标,得到了最终的综合性能评价模型。该研究以广东省广云高速为工程实例,确定了该高速公路路面使用性能5个评价指标的权重,建立了改进后的路面使用性能综合评价模型。最后将该评价方法与规范的评价方法进行对比分析,可以看出该模型比较符合实际情况。Pavement performance evaluation is an important part of pavement maintenance in China,and is of great significance to pavement maintenance management.In order to make the road performance evaluation more reasonable,based on the comprehensive evaluation indicators proposed by the specification,a new evaluation method is proposed in the paper.The BP neural network is used in the method,and at the same time,the structural strength index of the pavement is considered,and the final comprehensive performance evaluation model is obtained.In this study,Guangzhou-Yunfu Expressway is selected as the project example,and the weights of five evaluation indexes of Guangzhou-Yunfu Expressway’s pavement performance are determined.The improved comprehensive evaluation model of pavement performance is established.Finally,the evaluation method is compared with the standard evaluation method,from which it can be seen that the model is more in line with the actual situation.

关 键 词：路面养护 路面性能评价 BP神经网络 权重 

分 类 号：U416.2[交通运输工程—道路与铁道工程]