基于随机森林的隧道建成初期累计沉降量预测
投稿时间:2021-05-17  修订日期:2021-07-20  点此下载全文
引用本文:
摘要点击次数: 164
全文下载次数: 0
作者单位邮编
林广东* 中交一公局集团有限公司 100020
何军 中交隧道工程局有限公司 
申小军 中交隧道工程局有限公司 
徐龙飞 长安大学公路学院 
裴莉莉 长安大学信息工程学院 
余婷 长安大学信息工程学院 
基金项目:国家重点基础研究发展计划(973计划)
中文摘要:隧道建设是公路建设中一类重要的工程,沉降是隧道质量监测的一项重要指标。为预防沉降带来的安全隐患,确保后期隧道正常运行,在建成初期对隧道的沉降累积量进行实时监测并对后期沉降作出较为准确的预测是很有意义的。文章采用随机森林模型对隧道建成初期的隧道累计沉降量进行预测,并将该模型的预测结果与深度神经网络模型做对比分析。结果表明,随机森林模型对沉降累积量的预测精度更高,能够对隧道建成初期的累计沉降量进行有效的预测,为隧道的安全监测提供数据支持。
中文关键词:公路隧道  沉降预测  机器学习  随机森林  深度神经网络
 
Prediction of Cumulative Subsidence in the Initial Stage of Tunnel Construction Based on Random Forest
Abstract:Tunnel construction is an important project in highway construction. Settlement is an important index of tunnel quality monitoring. In order to prevent the hidden danger caused by subsidence and ensure the normal operation of the tunnel in the later period, it is of great significance to carry out real-time monitoring and make accurate prediction of the subsidence of the tunnel in the early stage of completion. The random forest model is used to predict the cumulative settlement of the tunnel in the early stage of construction, and compare it with the deep neural network model. The results show that the random forest model has higher accuracy in predicting the accumulation of subsidence. And it can effectively predict the accumulative subsidence at the early stage of tunnel construction, and provide data support for tunnel safety monitoring.
keywords:Highway tunnel  Subsidence prediction  Machine learning  Random forest  DNN
查看全文   查看/发表评论   下载pdf阅读器