New Roadbed Settlement Prediction Methods and Their Outlook
廖向阳LIAO Xiang-yang
（湖南省交通规划勘察设计院有限公司,长沙410000 )
(Hunan 'Transportation planning Survey and Design Institute Co.,Ltd.,Changsha 410000,China )
摘要:传统路基沉降预测方法非线性映射能力差。提出一种基于机器学习的路基沉降预测模型,并以江门市新会区银鹭大道路基工程为例,验证该预测模型的有效性。结果表明:所提预测模型被验证效果较好,各模型预测性能排序为:SVR模型>BP模型>RF模型;图形神经网络在路基沉降预测领域的应用具有良好发展前景。研究成果可为路基沉降的预测提供理论参考.
Abstract: 'The traditional roadbed settlement prediction method has poor nonlinear mapping capabilit.A machine leaning basedroadbed setlement prediction model is proposed and the effctiveness of the prediction model is verified by taking the moadbed project ofYinlu Avemue in Xinhui District, Jiangmen City as an example. The results showed that the proposed predtiction model was validated to beeffective, and the predliction performance of each model was ranked as: SVR model > BP model > RF model; the application of graphicalneural network in the field of roadbed setlement prediction has good development prospects. The research results can provide theoreticalreference for the prediction of roadbed settlement.
关键词:交通线路;路基沉降;沉降预测方法;机器学习
Key words: traffic line; roadbed settlement; settlement prediction method; machine learning
价值工程-文章出处：廖向阳.新型路基沉降预测方法及其展望[J].价值工程,2023,42(19):166-168. |
