Shield tunnel construction impacts old buildings in urban areas. Extensive building deformation data was collected for Tianjin Metro Line 7’s shield tunnel passing beneath old buildings using automated measurement robots. Machine learning algorithms were applied to analyze the correlation between building instantaneous settlement and shield construction parameters such as average velocity, thrust, grouting volume, shield distance, and grouting pressure. A predictive model for building settlement was established. The results show that old buildings within -50~70 m are affected by shield construction. Differential settlement is significant, with noticeable differences on various facades. Grouting volume, thrust, and average velocity positively correlate with instantaneous settlement, with shield distance having the greatest impact. Reasonable construction parameters ensure building settlement and deformation remain within acceptable limits. The machine learning-based predictive model closely aligns with actual settlement curves, demonstrating robust predictive capabilities. This provides valuable insights for predicting and controlling surface settlement in future shield tunnel projects.

With the emergence of deep learning technologies, speech enhancement methods based on deep learning have seen widespread application and generally surpass traditional approaches in performance. The fundamental framework of noise reduction signal processing in speech enhancement was outlined and progressively delved into the latest advancements in deep learning-driven speech enhancement models. A comprehensive organization of deep learning-based speech enhancement algorithms was provided, detailing the principles, characteristics, evaluation metrics, and representative studies of various neural network-based methods. The advantages and limitations of these approaches were thoroughly assessed. Finally, in light of the current developmental landscape, the core challenges encountered in the speech enhancement process were analyzed, and future developmental trajectories were discussed and predicted.

Due to the complex underground environment, low lighting conditions, and the small size of hard hats, the detection results are not ideal. To address low-quality images in complex environments, an improved YOLOv7 for hard hat detection in low-quality images from underground coal mines was proposed. Firstly, addressing the limitation that image features were susceptible to noise interference under low-light conditions, a multi-scale MELAN module was introduced. By constructing a multi-scale attention mechanism, broader contextual information was captured, thereby enhancing feature extraction and effectively suppressing noise interference. Secondly, the OD-SMP module was constructed using soft pooling and full-dimensional dynamic convolution in the backbone network, which reduced information diffusion in feature mappings, retained more contextual information, and enhanced the detection capability for small targets. Finally, to address the varying quality of detection samples caused by the complex backgrounds and environments with different lighting and distances in underground coal mines, Wise-IoU was used as the loss function. Experimental results show that the average precision of the improved model is 94.9%, which is 13.5% higher than the original YOLOv7 model, demonstrating better detection performance.

To enhance heat transfer efficiency and improve thermal exchange performance, a composite enhanced thermal exchange technology was explored that combined annular internal fins with protruding units, aiming to create an innovative thermal exchange structure. Through numerical simulation methods, the flow and heat transfer characteristics of this structure were studied within the Reynolds number Re range is 8 000~20 000. The analysis results indicate that the layout of the protruding units and four parameters (depth, radius, spacing, and quantity) have a significant impact on thermal performance. The mechanism of enhanced heat transfer was explained using field synergy theory. Under optimal parameters, with a depth of 2 mm, a radius of a specific value, a spacing of 20 mm, and six protruding units, the best thermal exchange performance is achieved, with an overall heat transfer performance improvement of 4.71%~23.59% compared to internal finned tubes. Increasing depth, radius, and quantity, while decreasing spacing, enhances heat transfer but also increases resistance, limiting the growth of overall thermal performance. Field synergy analysis shows that the structure promotes strong secondary vortices, significantly enhancing the synergy effect between the velocity field and the temperature field.

To improve the accuracy of precipitation forecasts and address the limitations of traditional numerical weather prediction models in forecast precision and computational efficiency, a meteorological large model was combined with a deep learning post-processing approach was combined. A case study was conducted for precipitation forecasts over Shaanxi Province during 2008—2018. Based on meteorological variable fields output by the FourCastNet model, a pre-trained model mapping meteorological fields to regional precipitation was constructed using Bayesian-optimized convolutional neural networks (CNN)/long short-term memory (LSTM) networks. The results indicate that this method outperforms traditional numerical weather prediction models in terms of spatial resolution and forecast accuracy. The regionally fine-tuned forecasts more accurately capture the spatiotemporal distribution of precipitation. Furthermore, the Bayesian-optimized deep learning post-processing algorithm effectively mitigates the impact of initial field biases on forecast results. These findings demonstrate the significant potential of integrating meteorological large models with deep learning post-processing algorithms for accurate precipitation forecasting, providing scientific support for disaster prevention, agricultural production, and water resource management.

In order to find out the uranium metallogenic potential of the Lower Cretaceous in Kelulun Sag of Hailar Basin, the spatial distribution of sequence stratigraphy, types and distribution characteristics of sedimentary facies and favorable spatial location of uranium enrichment in the area were studied by using sequence stratigraphy and sedimentary facies analysis methods based on core, logging and seismic data. The results show that the third-order and fourth-order sequence stratigraphic division marks of the target strata in the Kelulun Sag are determined. The target strata are divided into one super sequence, two three-order sequences and four fourth-order sequences, and the distribution characteristics of sequence stratigraphy are clarified. The sedimentary system of fan delta-lacustrine facies is mainly developed in the target layer in the area. The sedimentary facies plane has the characteristics of near source, fast phase change and small distribution range. The braided channel sand bodies of the fan delta plain and the underwater distributary channel sand bodies of the fan delta front have good uranium metallogenic potential.

To mitigate the risk of annular pressure buildup caused by solid-phase deposition in the B and C annuli of deep-water wells, experimental tests were conducted on sedimentation behavior using common deep-water drilling fluid systems. The sedimentation height and post-settling solid-phase permeability of various drilling fluids were measured. Based on the parameters of solid phase percolation characteristics, and considering the impact of annular fluid solid deposition, a predictive analytical method was established for annular pressure under percolation conditions. Case analysis was conducted to validate the approach. Results show that the sedimentation height follows the order: oil-based drilling fluid > EZFLOW drilling fluid > HEM drilling fluid. In contrast, the post-settling solid-phase permeability is ranked as EZFLOW drilling fluid > HEM drilling fluid > oil-based drilling fluid, with a maximum permeability of 2.216 μm². Under annular fluid solid-phase deposition conditions, reductions in annular fluid viscosity, increases in formation permeability, and longer open-hole cement sheath sections reduce fluid viscous resistance, enlarge the seepage contact area with the formation, and enhance fluid flow. Therefore, reducing drilling fluid viscosity and extending the open-hole cement sheath length can improve the pressure release capacity in the B and C annuli of deep-water wells. However, the presence of solid-phase deposition significantly restricts seepage flow rates compared to conditions without deposition, leading to a potential risk of incomplete pressure relief following solid-phase sedimentation.

With the development of urbanization, the number and scale of sewage treatment plants are increasing, and the effective collection of odor generated by them is of great significance to prevent odor leakage, protect the environment and reduce energy consumption. The influence of structural parameters (different tube bundle positions, different tube bundle numbers, different pipe diameters) and operating parameters (suction flow) on the internal flow field of the odor collection hood of a sewage treatment plant was studied by numerical simulation. The results show that when the tube bundle is placed at the very edge of the air collecting hood, the number of suction pipes is three, the diameter of pipe is 150 mm, and the suction flow range is 3.63 kg/s to 5.71 kg/s, the airflow structure in the cavity is the best, the odor concentration is the lowest, and the suction effect is the best.

Rainfall is one of the main factors affecting landslide stability. To explore the deformation and stability laws of unsaturated soil landslides under different rainfall conditions, based on the theory of saturation-unsaturation and the intensity reduction method, focused on a traction landslide in eastern Jiangxi. The rock and soil parameters of the landslide hazard body were determined through field investigations and laboratory tests. AutoCAD software was used to restore the terrain and geological conditions of the Tongshan landslide in Xishan Village, Zhengfang Town, as accurately as possible. A three-dimensional mathematical model of seepage-deformation coupling for the landslide was established. The dynamic process of seepage deformation and stability of the unsaturated soil landslide under various rainfall conditions (different intensities, durations, and post-rain stoppage) was simulated. The results indicate that the pore water pressure at the slope surface increases gradually with rainfall intensity and duration. The drainage velocity increases, the saturated zone gradually transforms into the unsaturated zone, and the displacement and deformation increase progressively. After the rainfall stops, leakage in the landslide hazard body exhibits a delayed response. The displacement deformation after continuous rainfall first increases and then decreases, while the displacement following intermittent rainfall shows periodicity during the short-term post-rain stoppage period. In terms of landslide stability, the reduction trend of the stability coefficient of the slope body strength under different rainfall intensities and durations is similar, but the stability coefficient under the same intensity varies, with the overall trend showing a gradual decrease with increased duration. After the rain stopped, the stability coefficient of the uniform rainfall slope body decreases initially and then increases, whereas the stability of the intermittent rainfall slope body exhibits periodic variations.

Groundwater plays a pivotal role in the production and sustenance of life. However, the potential geologic risks associated with its exploitation must be acknowledged. Changes in groundwater levels have been shown to precipitate geologic disasters such as landslides, mudslides, and ground subsidence. Therefore, the mastery of groundwater information is of great scientific significance for disaster prevention, mitigation, and the rational use of water resources. The temperature tracing method is recognized as a promising technique with significant applications in preventing and providing early warning of geological hazards, such as landslides and mudslides. Among the many methods available, this technique was noted for its great potential. The recent groundwater exploration methods, theoretical research, and new indoor experimental research methods was focused on. The latest research progress related to the groundwater method of geothermal inversion in the seepage of rivers and dams, landslides, and groundwater exploration was reviewed. Through comparative analysis with the traditional electric method of exploration, current theoretical models and new problems faced by the practice of engineering exploration were analyzed. Future research should focus on multi-field coupling, multi-parameter integration, analysis of groundwater patterns in special soil sites, dynamic monitoring of groundwater for major projects, and early warning and prediction of geological disasters will be focused on. These research directions will provide essential scientific and technical support for the prevention of geological disasters and water resources management.