Effect of Steel Support Cross-Section and Preloaded Axial Force on the Stability of Deep Foundation Pits

Buildings, Journal Year: 2024, Volume and Issue: 14(8), P. 2532 - 2532

Published: Aug. 16, 2024

To investigate the effects of steel support cross-section dimensions and preloaded axial force levels on stability foundation pits, numerical simulations were conducted for open-cut deep pits based monitoring data from Changchun Metro Line 9. Results show that increasing wall thickness diameter significantly reduces horizontal displacement enclosure pile. When increases 14 mm to 25 mm, pile can be reduced by up 7.63 11.4–15%. changes 609 800 second is 3.2–5.5%. The change in results a about 3–5 surface settlement 0.6–4.2 mm. meets pit control requirements when it reaches 60% design force.

Language: Английский

---

Machine Learning-Based 3D Soil Layer Reconstruction in Foundation Pit Engineering

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(8), P. 4078 - 4078

Published: April 8, 2025

In the construction of deep foundation pits, early warning measures are essential to reduce risks and prevent personnel injuries. underground structure pressure analysis, soil layer support data indispensable. Therefore, reconstruction serves as a critical step, while sparse borehole limit accuracy traditional methods. This paper proposes machine learning-based method address this issue. First, various types generated by simulating formation process Earth’s layers, thereby providing sufficient training data. Subsequently, coding algorithm is designed extract features inputs for convolutional neural network. Finally, 3D meshing performed on from real boreholes, model rendering achieved through voxel clustering algorithm. The an rate over 90% in tests demonstrated excellent robustness. By applying algorithm, we successfully reconstructed layers at typical pit site Chinese city, validating its effectiveness real-world scenarios potential large-scale engineering applications.

Language: Английский

---

Bridge management with AI, UAVs, and BIM

Automation in Construction, Journal Year: 2025, Volume and Issue: 175, P. 106170 - 106170

Published: April 9, 2025

Language: Английский

---

Time-series prediction of the deformation behavior of small circular ring based on residual structure and Bi-LSTM network

Robotic Intelligence and Automation, Journal Year: 2025, Volume and Issue: unknown

Published: April 26, 2025

Purpose The purpose of this paper is to propose a novel deep learning approach accurately predict the time-series deformation behavior and load-displacement response curves small circular ring specimens under different displacement rates, using existing experimental data. This method aims reduce costs improve efficiency compared traditional testing methods. Design/methodology/approach A network model based on residual structure Bidirectional Long Short-Term Memory established. tensile test data from open set used as input. Data preprocessing performed, including time point selection standardization. Mean squared error loss function, with classical MLP, models. Findings For nine samples, proposed achieves an R-squared 0.9982 mean absolute 23.6810, outperforming other five models, which had values ranging 0.9805 0.9970 between 26.3013 96.7325. Compared demonstrates excellent stability accuracy in initial final stages prediction. Originality/value research pioneered application techniques field testing, proposing data-driven distinct combines convolutional networks, fully connected layers LSTM into hybrid architecture capable effectively analyzing methods requiring extensive uses limited data, reducing improving studying behavior. It provides new ideas for digital transformation intelligent analysis materials science.

Language: Английский

---

Foundation Pit Support Deformation Prediction Based on Deep Echo State Network Optimized by Empirical Balance Particle Swarm Optimization

Published: Sept. 13, 2024

Language: Английский

---