Prediction of Shield Machine Attitude Based on Various Artificial Intelligence Technologies

The shield machine attitude (SMA) is the most important parameter in the process of tunnel construction. To prevent the shield machine from deviating from the design axis (DTA) of the tunnel, it is of great significance to accurately predict the dynamic characteristics of SMA. We establish eight SMA prediction models based on the data of five earth pressure balance (EPB) shield machines. The algorithms adopted in the models are four machine learning (ML) algorithms (KNN, SVR, RF, AdaBoost) and four deep learning (DL) algorithms (BPNN, CNN, LSTM, GRU). This paper obtains the hyperparameters of the models by utilizing grid search and K-fold cross-validation techniques and uses EVS and RMSE to verify and evaluate the prediction performances of the models. The prediction results reveal that the two best algorithms are the LSTM and GRU with EVS > 0.98 and RMSE < 1.5. Then, integrating ML algorithms and DL algorithms, we design a warning predictor for SMA. Through the historical 5-cycle data, the predictor can give a warning in advance if the SMA deviates significantly from DTA. This study indicates that AI technologies have considerable promise in the field of SMA dynamic prediction.

Study on the Mechanical Characteristics of the Screw Driving Torque during the Emptying of the Shield Soil Bin

This article studies the changing law of the driving torque of the screw conveyor during the emptying process of the earth pressure balance shield chamber. First, the discrete element method of discrete medium theory and 3D software SolidWorks were used to create the research object and the screw conveyor model, and then the model parameters were determined and calibrated through numerical calculations and indoor experiments. The final numerical calculation results show that: 1) the screw torque will drop in waves with the increase of the calculation time. When the screw conveyor rotates at 360 deg/s, the calculated screw torque fluctuation amplitude is small; 2) when the number of particles in the soil bin is reduced to a certain extent, the use of a higher screw speed to improve the “dumping soil effect” is of little significance; 3) the negative exponential function can be used to better fit the decrease of screw torque with time; 4) for the bulk medium, for a given particle size and screw structure, there is a suitable speed, so that the effect of “machine-soil collision” is small, and the torque change of the screw is relatively stable.

Evaluation of Lateral and Axial Deformation for Earth Pressure Balance (EPB) Tunnel Construction Using 3 Dimension Finite Element Method

Mass Rapid Transit Jakarta (MRTJ) phase 1 tunnel construction using the earth pressure balance method has been completed and surface settlement and lateral displacement data according to elevation and inclinometer readings has been collected to evaluate the effect of tunnel’s construction on surrounding infrastructure. Soil stratification along the research area, defined according to boring logs and soil parameters for the hardening soil model (HSM) and the soft soil model (SSM), was determined by optimization of stress-strain curve fitting between CU triaxial test, consolidation test and soil test models in the Plaxis 3D software. Evaluation of the result of surface settlement measurements using an automatic digital level combined with geodetic GPS for elevation and position control points showed that the displacement behavior was affected by vehicle load and stiffness of the pavement. Lateral displacement measurements using inclinometers give a more accurate result since they are placed on the soil and external influences are smaller than surface settlement measurement. The result of 3D finite element modeling showed that surface settlement and lateral displacement during TBM construction can be predicted using HSM with 2% contraction. SSM and the closed-form solutions of Loganathan and Poulos are unable to provide a good result compared to the actual displacement from measurements.