交通节能与环保2025，Vol.21Issue(5)：105-109,114,6.DOI:10.3969/j.issn.1673-6478.2025.05.021
基于机器学习的盾构推力预测研究
Research on Thrust Prediction of Shield Tunneling Based on Machine Learning
摘要
Abstract
There are many parameters for shield tunneling,among which the shield thrust has a significant impact on soil pressure balance and formation settlement.Effective prediction of shield tunneling thrust can guide construction and reduce engineering accidents,making it a hot research topic in intelligent shield tunneling.Based on the excavation data of the shield tunnel on the R2 line of Jinan Metro,a thrust prediction model was established using three algorithms:random forest,XGBoost,and BP neural network.The model was evaluated through cross validation,and the goodness of fit R2(R-Squared),mean square error(MSE),and mean absolute error(MAE)were selected as the evaluation indicators for the model.The comparative analysis between the predicted and measured values of shield tunneling thrust shows that the R2 values of the three algorithm models are all greater than 0.6,which can meet the engineering accuracy requirements.The XGBoost model has the best prediction effect.关键词
机器学习/盾构推力预测/随机森林/XGboost/BP神经网络Key words
machine learning/shield thrust force prediction/random forest/XGboost/BP neural network分类
交通工程引用本文复制引用
王文正,林雪冰,宋克志,师鸿儒,邵煜琪,邵长至..基于机器学习的盾构推力预测研究[J].交通节能与环保,2025,21(5):105-109,114,6.基金项目
国家自然科学基金项目(51978322) （51978322）
