The particle-size distribution (PSD) of rock fragments is crucial for various engineering practices. Recently, there has been increasing attention towards new methods for analysing rock fragment PSD based on artificial intelligence and computer vision. However, most existing image-based studies on this topic are case-based, while heavily relying on manual annotation. The proposed algorithms or trained models often exhibit poor performance across different geological and lighting conditions. In this research, an automatic scene-adaptive framework for real-time rock fragment recognition (FragAdaptixAI) is introduced. First, a diverse foundation image set of rock fragments is built, and an automated annotation module based on an improved segment anything model (SAM) is employed to create the foundation rock fragment dataset, which is then utilised for training the foundation model (RTMDet). When faced with new cases, the foundation model is rapidly fine-tuned using a small amount of current case data to derive an application model, supporting real-time automation in rock fragment recognition. FragAdaptixAI was tested using two cases, and the test results demonstrate that (1) the automated annotation module, developed based on the improved SAM, is effective, and (2) FragAdaptixAI achieves excellent results with short tuning time for new cases. The approach introduces a new concept to address the generalisation issues in rock fragment recognition by combining the zero-shot capability of large models with the real-time processing of small models.
This study examines the in-situ lateral static load behavior of a closed-diaphragm wall foundation, aiming to better understand its load–displacement response, structural behavior, and soil interaction under horizontal loading. An in-situ static load test was conducted with a maximum applied load of 70 MN, revealing that the diaphragm wall initially exhibits a linear load–displacement response, which becomes increasingly nonlinear as the load increases. The horizontal displacement of the lateral walls is nearly identical to the overall displacement of the diaphragm wall, making it a reliable indicator of the wall’s load state, particularly when it is challenging to measure total displacement. The wall behaves as a rigid body with minimal relative displacement between sections, and overturning failure is identified as the primary failure mode. Earth pressure distribution varies around the wall: passive earth pressure is observed at the front edge, while active and passive pressures alternate at the rear edge. These findings provide valuable insights into the design of diaphragm wall foundations, emphasizing the importance of lateral displacements.
Suction bucket foundations offer efficient installation and reusability for offshore wind turbines, yet their uplift behavior in sands with different permeabilities remains insufficiently understood. This study investigates the uplift behavior of suction bucket foundations in sands with varying permeabilities. Uplift capacity increases by up to 102% as permeability decreases from 7 × 10−2 to 7 × 10−3 cm/s, with a diminishing rate of improvement beyond k ≈ 10−4 cm/s. A recommended uplift rate range of 10−5–10−4 m/s was found to optimize performance, minimizing hydraulic demand and seabed disturbance. The study highlights the critical role of passive suction in enhancing uplift capacity under partially drained conditions and provides key insights for offshore foundation design.
Coal mining can cause the rupture of the overlying strata, and the energy released by large-scale fractures can therefore induce earthquake disasters, which in turn can cause more secondary disasters. In the past 50 years, countless earthquakes induced by coal mining have been reported. In this paper, the main factors relating to the mining-induced seismicity, including the mechanical properties, geometry of the space, excavation advance, and excavation rate, are investigated using both experimental and numerical methods. The sensitivity of these factors behaves differently with regard to the stress distribution and failure mode. Space geometry and excavation advances have the highest impact on the surface settlement and the failure, while the excavation rate in practical engineering projects has the least impact on the failure mode. The numerical study coincides well with the experimental observation. The result indicates that the mechanical properties given by the geological survey report can be effectively used to assess the risk of mining-induced seismicity, and the proper adjustment of the tunnel geometry can largely reduce the surface settlement and improve the safety of mining.