Clayshock mechanism and application to shield tunneling through existing tunnels: Settlement prediction using artificial intelligence

Abstract

In this study, we investigated the mechanism of clayshock during the construction of a new shield tunnel that passes through an existing shield tunnel. The experimental findings demonstrated a significant reduction of approximately 12% in the friction between the shield machine and the clay and silt layers. After the clayshock concentration exceeds 400 kg/m³, it may successfully impede the movement of the slurry, completely fill the space between the formation and the shield, and minimize the surface settling. Furthermore, Plackett–Burman testing was used to assess the sensitivity of the primary shield characteristics that affected settlement. The findings indicated that settlement was greatly influenced by the earth pressure, grouting quantity, and clayshock efficiency. A full factor analysis revealed an intricate link between the shield properties and settlement. If the volume of grouting was less than 6.5 m³, it was very difficult to regulate the settling value within −1.0 mm. We utilized the following neural network models: gated recurrent unit (GRU) network, long short-term memory network (LSTM), and Elman recurrent neural network (ELMAN). The training input data consisted of geometric, geological, and shield parameters, and the prediction output data served as the settlement value. Researchers will be able to generate accurate and reliable predictive outcomes by improving the model training process.

Keywords

• shield crossing
• clayshock mechanical experiment
• settlement prediction
• neural network