A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
A Dynamic Model of Machine Learning and Deep Learning in Shield Tunneling Parameters Prediction
Tunnel boring machine (TBM) is widely utilized in large cross-section tunnel excavation with high permeability soil conditions. During the shield tunneling process of TBM, dynamic prediction of driving parameters including advance rate and cutterhead torque is required. This study establishes a dynamic prediction model (DPM), LSTM-XGBoost, based on machine learning and deep learning algorithm framework incorporating with wavelet transform, long short-term memory method (LSTM) and extreme gradient boosting method (XGBoost). The proposed DPM can predict TBM-driving parameters in advance and adjust parameter weight in order to optimize operation. Moreover, the generalization ability of the proposed DPM is compared with traditional machine learning algorithms such as support vector machines (SVM) based on the data obtained in a practical tunneling project. The comparison results reveal that the LSTM-based DPM is suitable for prediction of time-series data.
A Dynamic Model of Machine Learning and Deep Learning in Shield Tunneling Parameters Prediction
Tunnel boring machine (TBM) is widely utilized in large cross-section tunnel excavation with high permeability soil conditions. During the shield tunneling process of TBM, dynamic prediction of driving parameters including advance rate and cutterhead torque is required. This study establishes a dynamic prediction model (DPM), LSTM-XGBoost, based on machine learning and deep learning algorithm framework incorporating with wavelet transform, long short-term memory method (LSTM) and extreme gradient boosting method (XGBoost). The proposed DPM can predict TBM-driving parameters in advance and adjust parameter weight in order to optimize operation. Moreover, the generalization ability of the proposed DPM is compared with traditional machine learning algorithms such as support vector machines (SVM) based on the data obtained in a practical tunneling project. The comparison results reveal that the LSTM-based DPM is suitable for prediction of time-series data.
A Dynamic Model of Machine Learning and Deep Learning in Shield Tunneling Parameters Prediction
Lecture Notes in Civil Engineering
Geng, Guoqing (editor) / Qian, Xudong (editor) / Poh, Leong Hien (editor) / Pang, Sze Dai (editor) / Wang, Ruohan (author) / Chen, Guan (author) / Liu, Yong (author)
Proceedings of The 17th East Asian-Pacific Conference on Structural Engineering and Construction, 2022 ; Chapter: 99 ; 1241-1254
2023-03-14
14 pages
Article/Chapter (Book)
Electronic Resource
English
Tunnel excavation , LSTM , XGBoost , Dynamic prediction , Intelligent shield tunnelling Engineering , Building Construction and Design , Structural Materials , Solid Mechanics , Sustainable Architecture/Green Buildings , Light Construction, Steel Construction, Timber Construction , Offshore Engineering
Prediction of shield tunneling-induced ground settlement using machine learning techniques
| Springer Verlag | 2019
Prediction of shield tunneling-induced ground settlement using machine learning techniques
| Springer Verlag | 2019