科技创新与应用2024，Vol.14Issue(30)：52-54,59,4.DOI:10.19981/j.CN23-1581/G3.2024.30.013
基于GPR模型的多孔沥青混合料空隙率预估
摘要
Abstract
The void content of porous asphalt mixture is one of the key indicators that affect its drainage function and road performance.In order to quickly determine the void content of porous asphalt mixtures,this study took the passage rate of different sieve size and oil-stone ratio of the mixture gradation as independent variables,extracted characteristic parameters through correlation analysis,and then established the PAC-13 porous asphalt mixture void content prediction model based on the Gaussian Process Regression(GPR)model,and compared and analyzed the accuracy of the GPR model with multiple linear regression,AdaBoost and random forest methods in predicting the void content of porous asphalt mixtures.The results show that the GPR model for predicting the porosity of porous asphalt mixtures using the sieve passage rates of 4.75 mm,2.36 mm,1.18 mm,0.6 mm,0.3 mm,0.15 mm and 0.075 mm and oil-to-stone ratio as model parameters has good accuracy,with a linear fitting coefficient of 0.95;compared with multiple linear regression,AdaBoost and random forest methods,the GPR model is relatively more applicable to predicting the porosity of porous asphalt mixtures.关键词
道路工程/多孔沥青混合料/空隙率/高斯过程回归/预估模型Key words
road engineering/porous asphalt mixture/void content/Gaussian Process Regression(GPR)/prediction model分类
交通运输引用本文复制引用
马志鹏,章启月,张泽霖,肖一帆,邓学耀,刘祥..基于GPR模型的多孔沥青混合料空隙率预估[J].科技创新与应用,2024,14(30):52-54,59,4.基金项目
云南省大学生创新创业训练计划项目(S202210674133) （S202210674133）
云南省教育厅科学研究基金项目(2023J0132) （2023J0132）
