基于变权组合的沥青混凝土路面损坏状况指数预测  

Condition index prediction of asphalt concrete pavement surface damage based on variable weight combination

作　　者：李海莲[1] 周思汝 李清华 刘忠磊 贾卫东 LI Hai-lian;ZHOU Si-ru;LI Qing-hua;LIU Zhong-lei;JIA Wei-dong(School of Civil Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China;China First Highway Engineering Co.,Ltd.,Beijing 100024,China;China Railway Fourteen Bureau Group Co.,Ltd.,Jinan 250101,China)

机构地区：[1]兰州交通大学土木工程学院,兰州730070 [2]中交一公局集团有限公司,北京100024 [3]中铁十四局集团有限公司,济南250101

出　　处：《兰州大学学报(自然科学版)》2025年第1期35-42,共8页Journal of Lanzhou University(Natural Sciences)

基　　金：国家自然科学基金项目(51868042);甘肃省自然科学基金项目(20JR10RA229,22JR5RA334);甘肃省高等学校创新基金项目(2021A-048);兰州交通大学“百名青年优秀人才培养计划”基金项目(2018103)。

摘　　要：以某高速公路若干路段为工程背景,运用灰色关联度分析确定影响路面破损的关键因素,基于GM(1,N)模型与BP神经网络模型对路面损坏状况指数进行预测,结合组合原理和赋权思想提出3种变权组合预测模型.通过误差检验对各模型予以综合评价,变权组合预测模型能充分利用各模型的优势,有效提高路面使用性能预测的准确性,并根据特定路段预测结果择优选取.4个路段的预测结果与实际值的均方根误差分别为3.21×10^(-5)、4.24×10^(-5)、2.13×10^(-5)、4.22×10^(-3).路面损坏状况作为路面使用性能评价的组成部分,需要准确把握路面损坏状况指数的发展趋势.Several sections of an expressway were taken as the engineering background,and the grey correlation analysis used to determine the key factors affecting the pavement damage.Based on the GM(1,N) model and BP neural network model,the pavement surface condition index (P_(CI)) was predicted,and three variable weight combination prediction models were proposed based on the combination principle and weighting idea.Through an error test,a comprehensive evaluation of each model showed that the variable weight combination prediction model could make full use of the advantages of each model,effectively improve the accuracy of pavement performance prediction,and select the best according to the prediction results of specific road sections.The results showed that the root mean square error between the predicted results and the actual values of the four sections was 3.21×10^(-5),4.24×10^(-5),2.13×10^(-5) and 4.22×10^(-3),respectively.As part of pavement performance evaluation,pavement damage status needs to accurately grasp the development trend of P_(CI) and thus provide data support for expressway pavement maintenance management.

关 键 词：灰色关联度分析 沥青混凝土路面 GM(1 N)模型 路面损坏状况指数 变权组合模型 

分 类 号：TU416[建筑科学—岩土工程]