A critical evaluation of machine learning and deep learning in shield-ground interaction prediction

Zhang, Pin; Wu, Huai-Na; Chen, Renpeng; Tian, Dai; Meng, Fanyan; Wang, Hongbo

doi:10.1016/j.tust.2020.103593

Cited by 97 publications

(21 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gao et al [20] adopted three different recurrent neural network (RNN) models to predict TBM thrust and thrust in real-time based on in-situ operating data. Zhang et al [21] found that LSTM model is better suitable for predicting the shield load than random forest (RF) model. Chen et al [22] predicted torque and thrust based on an improved LSTM algorithm, and making it possible to adjust the TBM tunneling parameters in real time.…”

Section: B Intelligent Prediction For Shield Loadmentioning

confidence: 99%

A Novel Hybrid Deep Neural Network Prediction Model for Shield Tunneling Machine Thrust

Chen

Jia

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

Section: B Intelligent Prediction For Shield Loadmentioning

confidence: 99%

A Novel Hybrid Deep Neural Network Prediction Model for Shield Tunneling Machine Thrust

Chen

Jia

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Many scholars have studied the ground settlement caused by shield construction. Table 1 shows the commonly used research methods [2][3][4][5][6][7][8][9][10][11][12][13]. The machine learning method is based on the existing monitoring data, which has high-accuracy calculation results, a short calculation time, and high timeliness.…”

Section: Introductionmentioning

confidence: 99%

A Study of Prediction of Ground Settlement of Shield Tunnel Based on EEMD-BO-GRU Algorithm

Xu²,

Yang³

et al. 2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Because the finite element method cannot meet the needs of shield tunnel engineering, multiple linear regression and traditional machine learning algorithms are difficult to effectively solve the shield tunnel’s nonlinear and non-stationary ground surface settlement data. Based on the Gated Recurrent Unit (GRU) algorithm of Ensemble empirical mode decomposition (EEMD) and Bayesian optimization (BO), this paper puts forward a prediction method of ground surface settlement during shield tunneling. Taking the shield tunnel construction of Zhijiang Road in Hangzhou as an example, by comparing with the traditional machine learning algorithm, the results show that RMSE, MAE, and MAPE of this algorithm are increased by 1.9313 mm, 1.7526 mm, and 4.9553%. It can be explained that this method has higher prediction accuracy and a more stable prediction effect, and the EEMD algorithm can effectively improve the generalization ability of the algorithm.

show abstract

“…In recent years, the use of data-based techniques such as machine learning (ML) to predict geotechnical issues has received increasing attentions [26,64,68,72]. Of numerous emerging ML techniques, artificial neural networks (ANN) is the most common approach to develop forecasting models of various geotechnical issues such as slope stability [40], soil properties [29,48], bearing capacity [7,28], deep excavation [2,71], mining [12,57], tunnelling [59], jet grouting [58], among others.…”

Section: Introductionmentioning

confidence: 99%

Base resistance of super-large and long piles in soft soil: performance of artificial neural network model and field implications

2022

View full text Add to dashboard Cite

This study aims to examine the performance of artificial neural network (ANN) model based on 1137 datasets of super-large (1.0–2.5 m in equivalent diameter) and long (40.2–99 m) piles collected over 37 real projects in the past 10 years in Mekong Delta. Five key input parameters including the load, the displacement, the Standard Penetration Test value of the base soil, the distance between the loading point and pile toe, and the axial stiffness are identified via assessing the results of field load tests. Key innovations of this study are (i) use of large database to evaluate the effect that random selection of training and testing datasets can have on the predicted outcomes of ANN modelling, (ii) a simple approach using multiple learning rates to enhance training process, (iii) clarification of the role that the selected input factors can play in the base resistance, and (iv) new empirical relationships between the pile load and settlement. The results show that the random selection of training and testing datasets can affect significantly the predicted results, for example, the confidence of prediction can drop under 80% when an average R2 > 0.85 is required. The analysis indicates predominant role of the displacement in governing the base resistance of piles, providing significant implication to practical designs.

show abstract

A critical evaluation of machine learning and deep learning in shield-ground interaction prediction

Cited by 97 publications

References 63 publications

A Novel Hybrid Deep Neural Network Prediction Model for Shield Tunneling Machine Thrust

A Novel Hybrid Deep Neural Network Prediction Model for Shield Tunneling Machine Thrust

A Study of Prediction of Ground Settlement of Shield Tunnel Based on EEMD-BO-GRU Algorithm

Base resistance of super-large and long piles in soft soil: performance of artificial neural network model and field implications

Contact Info

Product

Resources

About