摘要
为了准确预测沥青路面使用性能的变化趋势,针对道路路面性能数据积累年限少、道路检测单元数据波动大的特点,通过熵值赋权对多个检测单元路段数据进行赋权提取,结合传统灰色模型(GM(1,1))进行预测,并考虑到路面在服役期内,由于荷载、材料和环境等不确定性因素导致的使用性能的变化,建立马尔可夫模型进行误差修正,得到了兼顾智能、概率的熵值赋权灰色马尔可夫复合模型。结合江西省某国道检测单元K1580~K1587段路面损坏状况指数(pavement condition index,PCI)、路面行驶质量指数(riding quality index,RQI)数据进行预测验证,通过计算预测结果与实测值的方差发现,相比于传统的灰色预测模型和标准路面性能指数(pavement performance index,PPI)模型,熵值赋权灰色马尔可夫预测值与实测值间的方差更小,预测结果更具有代表性,预测精度更高,可为路面养护人员制定养护决策提供高可靠性分析。
In order to accurately predict the change trend of asphalt pavement performance,in view of the characteristics of less accumulation years of road pavement performance data and large fluctuation of road detection unit data,the entropy weighting was used to extract the weight of multiple detection unit road data.And the traditional gray model(GM(1,1))was used for the performance prediction.Considering the changes of uncertain factors in use performance such as load,material and environmental variability,the Markov model was established for error correction,and the entropy weighted grey Markov compound model considering intelligence and probability was obtained.The prediction verification was performed with the pavement condition index(PCI)and riding quality index(RQI)data of K1580-K1587 section of a national highway detection unit in Jiangxi province.Compared with the gray prediction model and standard pavement performance index(PPI)model,the predicted value and measured value of entropy weighted gray Markov model has the smallest variance.The prediction result is more representative,and the prediction accuracy is higher,which can provide reference for road maintenance personnel to make maintenance decisions analysis.
作者
蒋玮
许庆正
单金焕
胡勇
肖晶晶
张博文
袁东东
JIANG Wei;XU Qingzheng;SHAN Jinhuan;HU Yong;XIAO Jingjing;ZHANG Bowen;YUAN Dongdong(School of Highway, Chang’an University, Xi’an 710064, China;Jiangxi Highway Engineering Testing Center, Nanchang 330100, China;School of Civil Engineering, Chang’an University, Xi’an 710061, China)
出处
《中国科技论文》
CAS
北大核心
2022年第6期595-601,共7页
China Sciencepaper
基金
江西省交通运输厅科技项目(2018Z0001)。
关键词
道路工程
熵值赋权
灰色预测
马尔可夫误差修正
路用性能
road engineering
entropy weighting
grey prediction
Markov error correction
pavement performance
