摘要
随着城市化进程的加快和交通运输需求的增长,确保沥青路面的使用性能对交通安全和经济发展至关重要。然而,如何准确评价沥青路面使用性能、预测潜在问题并采取及时的维护措施,一直是道路工程领域迫切需要解决的核心问题。因此,提出准确且高效的沥青路面使用性能评价及预测方法成为交通运输领域的重要任务。全面综述了机器学习方法在沥青路面使用性能评价及预测领域的研究进展,分析了支持向量机、随机森林、神经网络以及遗传算法等模型的基本原理、优缺点以及在实际工程中的应用情况。通过比较和分析这些方法,了解各种方法在沥青路面使用性能评价及预测方面的优点以及所面临的挑战。最后,提出了未来的研究方向和建议,以期为沥青路面的管理和养护提供更有效的理论支持和方法,以提高道路交通的安全性和可靠性。
With the acceleration of urbanization and the growth of transportation demand,it is very important to ensure the performance of asphalt pavement for traffic safety and economic development.However,how to accurately evaluate the performance of asphalt pavement,predict potential problems and take maintenance measures in time has always been the core problem that needs to be solved urgently in road engineering.Therefore,it is an important task to put forward accurate and efficient evaluation and prediction methods of asphalt pavement performance.In this paper the research progress of machine learning method in the field of asphalt pavement performance was comprehensively summarized,the basic principles,advantages and disadvantages of support vector machine,random forest,neural network and genetic algorithm were analyzed,as well as the application in practical engineering.The advantages and challenges of various methods in asphalt pavement performance evaluation and prediction were understood by comparison and analysis.Finally,the future research directions and suggestions are put forward to provide more effective theoretical support and methods for the management and maintenance of asphalt pavement,so as to improve the safety and reliability of road traffic.
作者
陈祎
杜政阳
黄渝淇
邬昌健
Chen Yi;Du Zhengyang;Huang Yuqi;Wu Changjian(School of Road and Bridge,Zhejiang Institute of Communications,Hangzhou 311112,China;School of Civil and Transportation Engineering,Hebei University of Technology,Tianjin 300401,China)
出处
《市政技术》
2024年第8期93-101,共9页
Journal of Municipal Technology
基金
浙江交通职业技术学院2023年第二批学校高层次人才引进专项课题。
关键词
道路工程
沥青路面
使用性能评价
支持向量机
随机森林
神经网络
遗传算法
road engineering
asphalt pavement
performance evaluation
support vector machine
random forest
neural networks
genetic algorithm
