Xinyu Jia

Monitoring data from deep foundation pits exhibit significant nonlinear, nonstationary, and spatiotemporal coupling characteristics. Traditional methods struggle to simultaneously characterize their spatial correlations and temporal evolution patterns. To address these issues, on the basis of measured data from a typical deep foundation pit project in Beijing, this paper proposes a spatiotemporal feature-fused CNN-LSTM-Transformer prediction model for support pile deformation. By constructing a spatiotemporal matrix of monitoring data, the model achieves the synergistic fusion of spatial feature extraction, temporal dependency modeling, and global correlation perception. The results of the comparative analysis indicate that the proposed model demonstrates stable predictive performance across different pile locations and operating conditions. Its root mean square error (RMSE) and mean absolute error (MAE) are reduced by approximately 20% compared with those of the CNN and CNN-LSTM models. Particularly in areas with severe deep deformation and during high-fluctuation stages, the model effectively mitigates prediction lag and error accumulation, demonstrating a superior response capability to local abrupt changes. The findings suggest that this method can provide a reliable data-driven approach for the dynamic prediction of support structure deformation during deep foundation pit construction.

With the rapid expansion of underground rail transit construction in China, the high carbon emissions associated with subway tunnels and stations have become an increasing concern. This study systematically examines the carbon emissions of prefabricated concrete–filled steel pipe columns (PCSPCs) during the construction phase of a Beijing subway station built via the pile beam arch (PBA) method, applying the life cycle assessment (LCA) methodology as a case study. An analytical framework for the synergistic optimization of carbon emissions and costs was developed. By systematically adjusting key design parameters—such as the column diameter, wall thickness, and concrete strength—it was possible to minimize both carbon emissions and project costs while meeting the ultimate load-bearing capacity requirements. The results indicate that the production phase of PCSPCs accounts for as much as 98.845% of total carbon emissions, with labor, materials, and machinery contributing 10.342%, 88.724%, and 0.934%, respectively. A sensitivity analysis revealed that steel plates have the greatest impact on carbon emissions, followed by steel reinforcement, whereas concrete and cement exhibit relatively lower sensitivities. The ultimate load-bearing capacity of PCSPCs increases with larger column diameters, thicker walls, and higher concrete strength grades, with the relationships displaying a nonlinear trend. The damage modes and performance of PCSPCs under different design parameters were further verified through finite element analysis. On the basis of the optimization algorithm used to adjust the design parameters, the carbon emissions and costs of the PCSPCs were reduced by 10.32% and 21.55%, respectively, while still meeting the load-bearing capacity requirements.