Search Results (26)

Due to the cross-section characteristics of rectangular pipe jacking, the “soil-carrying effect” of overlying soil migration with the pipeline is prone to occur during jacking in shallow strata, resulting in a sharp increase in jacking resistance and large deformation of the strata. In this paper, a visual similarity model test of the soil-carrying effect is carried out for shallow buried large-section rectangular pipe jacking. The experiment innovatively combines VIC-3D digital image correlation technology, a 3D laser scanner and a thin-film pressure sensor to monitor the displacement of deep soil, surface heave and pipe resistance in an all-round and high-precision way. The influence of the overburden ratio and pipe–soil friction coefficient on the evolution of back soil was systematically studied. The results show that the evolution of the soil-carrying effect presents the three-stage characteristics of ‘elasticity-slip-strengthening’, and the smaller the overburden ratio, the larger the friction coefficient. And the smaller the critical displacement of the back soil, the more severe the formation disturbance. Based on the principle of mechanical balance, this paper puts forward the theoretical prediction model of the whole soil-carrying effect, deduces the critical friction coefficient and the critical jacking mileage, and compares it with the experimental results, which provides a scientific basis for the optimization of construction parameters and safety control of shallow buried rectangular pipe jacking. Full article

Intelligent sheep behavior monitoring is vital for modern husbandry, but faces severe challenges in natural pastures due to high-density flock occlusion. Traditional 2D skeleton-based networks often suffer from depth ambiguity and feature collapse, misclassifying static tremors as dynamic displacement. To overcome this, we propose a robust multimodal dual-stream framework using skeletal and local visual fusion. The architecture features an upstream spatial perception stage utilizing YOLOv11m-Pose. To reduce annotation costs and improve robustness, we introduce an Active Hard-Example Mining mechanism, explicitly retaining difficult samples with severe overlapping or edge truncation. For downstream behavioral decisions, a multimodal dual-stream architecture processes the targets. The Spatio–Temporal Kinematic Stream employs a Kinematic Denoising Engine, incorporating a 1D Gaussian filter and displacement dead-zone gate to purify 2D coordinates before feeding them into a BiLSTM network. Concurrently, the Spatial Visual Stream uses a ResNet-50 backbone on cropped RGB patches to capture essential spatial context, addressing the limitations of pure coordinates. Finally, a weighted Softmax layer integrates both streams. Experiments on a complex real-world dataset validate this approach. A baseline kinematic-only model achieved just 69.05% overall accuracy and 68.18% walking precision. In contrast, our dual-stream fusion network achieved 93.26% overall accuracy, elevating walking precision to 97.14% and the eating F1-score to 94.29%. By effectively decoupling similar static and dynamic behaviors, this study demonstrates the indispensability of local visual features, establishing a high-precision baseline for smart livestock monitoring. Full article

To better address the accuracy bottleneck of RGB-only Camellia oleifera C.Abel fruit detection in complex orchard environments, this paper proposes CIAFNet—a dual-stream RGB-D fusion detection network—and evaluates its potential as a visual front end for relative 3D localization using sensor-measured depth and camera back projection under controlled conditions. With RGB images and depth maps as parallel dual-branch inputs, the network integrates the C3k2_PartialNetBlock for efficient intra-modal feature extraction with reduced computational redundancy, devises the cross-modal interaction and difference-aware adaptive fusion (CIDAF) module for adaptive cross-modal feature fusion, and adopts an SC-EUCB-augmented BiFPN in the neck to optimize multiscale feature aggregation and detail restoration during upsampling. Pseudo-depth maps generated from natural orchard RGB images via Depth Anything V2 were paired with RGB counterparts to build an RGB–pseudo-depth dataset. Synchronized RGB-D data collected by an Intel RealSense D435i under controlled conditions were used to quantify pseudo-to-sensor depth discrepancies and evaluate input adaptability. On the natural orchard test set, CIAFNet achieved 93.33% mAP@0.5 with only 10.49 GFLOPs and 3.80 M parameters. Second-stage fine-tuning improved CIAFNet’s adaptation to D435i-measured depth under controlled conditions. In the subsequent relative displacement consistency experiment, the mean absolute consistency errors along the X, Y, and Z axes were 3.10, 3.15, and 3.27 mm, respectively, and the mean 3D Euclidean consistency error was 5.60 mm. These results demonstrate that CIAFNet improves Camellia oleifera fruit detection using natural orchard RGB–pseudo-depth data and has potential as a visual front end for relative 3D localization under controlled conditions. Full article

Recent context-aware streaming 3D reconstruction frameworks provide a promising solution for online vehicle perception by maintaining anchor references, local pose windows, and trajectory memory. However, directly applying such frameworks to intelligent rail vehicles remains challenging because rail transit scenes are dominated by long straight motion, low-parallax visual observations, repetitive trackside structures, weak textures, and illumination variations. These characteristics may cause redundant context accumulation, unstable frame registration, and gradual trajectory drift. To address this problem, this paper proposes a motion-aware geometric context adaptation method for streaming 3D reconstruction of intelligent rail vehicles in low-parallax scenes. Instead of requiring task-specific large-scale retraining, the proposed method adapts the inference-stage geometric context using scale-normalized visual motion cues, including scale-normalized translational displacement, turning tendency, and inter-frame viewpoint variation. A motion-aware keyframe selection strategy suppresses redundant low-parallax frames while preserving geometrically informative observations in curved or pose-changing segments. An adaptive local pose reference window further regulates recent visual context to improve frame registration consistency. Experiments on rail transit sequences and the Oxford Spires dataset show that the proposed method achieves lower trajectory error than LingBot-Map and VIPE, while reducing redundant keyframe storage and preserving the qualitative continuity of rail-related structures. The method provides a practical motion-aware streaming 3D perception solution for rail transit inspection and digital infrastructure management. Full article

Freehand three-dimensional (3D) ultrasound enables cost-effective volumetric breast imaging, but accurate reconstruction requires reliable probe tracking during manual scanning. This study proposes a portable freehand 3D ultrasound framework using dual-rotary-encoder two-degree-of-freedom (2DoF) pose sensing to measure probe displacement and inclination during breast scanning. A slip-resistant roller mechanism and time-aware trajectory modeling were introduced to improve measurement robustness under practical scanning conditions. The framework was evaluated through robotic experiments and phantom-based volumetric reconstruction. Positional displacement experiments achieved root mean square errors (RMSEs) of 0.38 mm on dry surfaces and 0.81 mm under gel-coated conditions. Inclination sensing using the rotary encoder outperformed an inertial measurement unit (IMU), achieving an RMSE of 2.76° with improved temporal stability. Reconstruction experiments using a breast phantom with spherical inclusions demonstrated successful volumetric visualization across multiple scanning trajectories. Statistical analysis revealed significant effects of inclusion size and scanning trajectory on relative reconstruction error, as well as a significant interaction between the two factors. Larger inclusions generally exhibited lower relative errors, while the influence of scanning trajectory depended on the target size. These findings support the feasibility of the proposed reduced-dimensional mechanical pose sensing approach for reliable freehand 3D ultrasound reconstruction with reduced hardware complexity. Full article

Low-permeability reservoirs at the high-water-cut stage commonly suffer from dominant water channel development, poor sweep of weakly connected zones, and inefficient mobilization of remaining oil. Existing profile control or oil displacement agents can improve either flow diversion or microscopic oil displacement, but their single-agent evaluation does not fully explain the coupled process of sweep expansion and remaining oil mobilization. To address this issue, this study focuses on a previously optimized HK-0417/ALT-603 composite system and investigates its synergistic behavior at pore, core, and well group scales. Microscopic visualization displacement experiments were used to identify streamline redistribution and remaining oil evolution. Natural core experiments were conducted to evaluate injectivity adaptability and plugging persistence. Under slug injection conditions, the Box–Behnken design was employed to optimize the injection parameters. Finally, the field pilot response was analyzed based on production data from test wells in the Changqing Oilfield. The results show that the combination system simultaneously achieves streamline expansion and residual oil reduction: the injected fluid is redistributed toward weakly swept zones, large continuous oil bodies are fragmented and dispersed, and both sweep efficiency and oil displacement efficiency are superior to those of individual agents. Natural core experiments indicate that the injection pressure difference is generally controllable in cores with permeabilities ranging from 1.76 to 7.02 mD, and the plugging rate during subsequent water flooding reaches 75.47–80.54%. Response surface optimization yields the following optimal parameter combination: profile control slug volume = 0.41 pore volume (PV), oil displacement slug volume = 0.61 PV, injection rate = 0.19 mL/min, with a corresponding predicted enhanced oil recovery (EOR) of 18.52%. In the field pilot, the cumulative injection volumes of the two injectors are 41,898 kg and 61,472 kg, respectively. The injection pressure in the well group increases from 5.8 MPa to 7.0 MPa, the comprehensive water cut decreases from 90.6% to 85.3%, and the monthly decline rate is reduced from 0.5% to 0.2%. The proposed system mainly acts by increasing flow resistance and redirecting flow in high-water-cut channels, while it enhances oil detachment through interfacial tension reduction in oil-bearing pores. After optimizing the slug parameters, the field pilot exhibits a clear phased response and promising application potential. Full article

Paleogeomorphology exerts first-order control on the distribution of structural hydrocarbon reservoirs across regional unconformities, whereas variations in pore-throat architecture and flow capacity among different geomorphic units further govern hydrocarbon migration pathways and accumulation sites. Therefore, high-resolution reconstruction of regional paleogeomorphology is essential for effective exploration. This study investigates the Yanwu area of the Ordos Basin, where pre-Jurassic paleogeomorphology was reconstructed based on detailed stratigraphic analyses of the Yan’an Formation and the Yan-10 oil-bearing interval, and its influence on reservoir formation was systematically evaluated. Paleogeomorphology was delineated using well-log-based compensated impression methods integrated with localized 3D seismic inversion. Reservoir samples from distinct geomorphic units were analyzed through thin-section petrography, FESEM imaging, high-pressure mercury intrusion, and visualized micro-scale hydrocarbon charging experiments to characterize pore-throat systems and flow behavior. Four geomorphic units—paleohighs, slope zones, terraces, and valleys—were identified. Seismic inversion across the Yanwu tributary valley and the Honghe paleovalley confirms the reliability of the reconstructed geomorphology. Reservoirs within slope zones and terraces exhibit superior pore-throat structures, dominated by intergranular and dissolution pores, and display grid-like displacement patterns with higher ultimate recovery in micro-charging tests. Portions of the paleohighs show comparable reservoir quality and flow capacity. Results indicate that slope zones and terraces represent the most favorable hydrocarbon accumulation domains. Where overlying strata provide effective sealing, hydrocarbons preferentially accumulate on structural highs within these geomorphic units; in contrast, insufficient sealing transforms them into efficient migration conduits. Certain paleohighs may also host structural-high accumulations when capped by effective traps. The clarified accumulation patterns across geomorphic units offer a robust framework for guiding hydrocarbon exploration and reserve growth in regions with similar tectono-sedimentary settings. Full article

Enhanced oil recovery (EOR) in fractured reservoirs presents significant challenges due to fluid channeling and poor sweep efficiency. In this study, a synergistic EOR system was developed with polymer-based weak gel as the primary component and foam as the auxiliary enhancer. The system utilizes a low-concentration polymer (1000 mg·L⁻¹) that forms a weakly cross-linked three-dimensional viscoelastic gel network in the aqueous phase, inheriting the core functions of viscosity enhancement and profile control from polymer flooding. Foam acts as an auxiliary component, leveraging the high sweep efficiency and strong displacement capability of gas in fractures. These two components synergistically create a multiscale enhancement mechanism of “bulk-phase stability control and interfacial-driven displacement.” Systematic screening of seven foaming agents identified an optimal formulation of 0.5% SDS and 1000 mg·L⁻¹ polymer. Two-dimensional visual flow experiments demonstrated that the polymer-induced gel network significantly improves mobility control and sweep efficiency under various injection volumes (0.1–0.7 PV) and gravity segregation conditions. Numerical simulation in a 3D fractured network model confirmed the superiority of this enhanced system, achieving a final oil recovery rate of 75%, significantly outperforming gas flooding (65%) and water flooding (59%). These findings confirm that weakly cross-linked polymer gels serve as the principal EOR material, with foam providing complementary reinforcement, offering robust conformance control and enhanced recovery potential in fracture-dominated reservoirs. Full article

To address the issues of easy degradation, dehydration, and insufficient deep plugging strength of traditional pre-crosslinked gel particles (PPGs) in high-temperature and high-salinity reservoirs, this study innovatively introduced amphiphilic carbon dots (CDs) with both hydrophilic and hydrophobic structures as multifunctional modifiers. The carbon dot-reinforced PPGs (CD-PPGs) were successfully prepared through in situ polymerization. Through systematic characterization, microscopic visualization experiments, and macroscopic oil displacement evaluation, the performance enhancement mechanism and profile control behavior were deeply explored. The results show that the amphiphilic carbon dots significantly enhanced the material’s temperature resistance (up to 110 °C), salt resistance (up to 15 × 10⁴ mg/L salinity), and mechanical properties by constructing a “hydrogen bond-hydrophobic association” dual crosslinking system within the PPG network. More importantly, it was found that CD-PPGs exhibit a unique “self-aggregation” ability in deep reservoirs, which enables the in situ formation of high-strength plugging micelles at the target location while ensuring excellent injectability. At a permeability range of 539.0–2988.6 mD, the sealing rate of 0.5 PV CD-PPGs was greater than 95%. With permeabilities of 490.1 mD and 3020.5 mD under heterogeneous reservoir simulation conditions, the total recovery degree after the CD-PPGs was 52.6%, which was 20.5% higher than that of single water flooding. This study not only developed a high-performance profile control nanomaterial but also elucidated its strengthening mechanism, providing new insights and a theoretical basis for advancing deep profile control technology. Full article

Well pattern infill adjustment combined with chemical flooding is an important technical approach for significantly improving oil recovery in high-water-cut reservoirs after polymer flooding. Current research predominantly focuses on the evaluation of oil displacement potential through either well pattern infilling or chemical flooding alone, while systematic experimental investigations and mechanism studies on the synergistic effect of well pattern infilling and chemical flooding remain insufficient. To overcome the limitations of single adjustment measures, this study proposes a synergistic improved oil recovery (IOR) strategy integrating branched preformed particle gel (B-PPG) heterogeneous phase composite flooding (HPCF) with well pattern infill adjustment. Two-dimensional visual physical simulation experiments are conducted to evaluate the synergistic oil displacement effects of different displacement systems and well pattern adjustment strategies after polymer flooding and to elucidate the synergistic IOR mechanisms under the coupling of dense well patterns and chemical flooding. The experimental results demonstrate that, under well pattern infill conditions, the HPCF system exhibits significant water control and oil enhancement effects during the chemical flooding stage, achieving a 29.95% increase in stage recovery compared to the water flooding stage. The system effectively blocks high-permeability channels while enhancing displacement in low-permeability zones through a coupling effect, thereby significantly expanding the displacement sweep volume, improving displacement uniformity, and efficiently mobilizing the remaining oil in low-permeability and residual oil-rich areas. Meanwhile, well pattern infill adjustment optimizes the injection–production well pattern layout, shortens the inter-well spacing, and effectively increases the displacement pressure differential between injection and production wells. This induces disturbances and reconfiguration of the streamline field, disrupts the original high-permeability channel-dominated flow regime, further expands the sweep range of the remaining oil, and substantially improves overall oil recovery. The findings of this study enrich and advance the theoretical framework of water control and potential tapping, as well as synergistic IOR mechanisms, in high-water-cut and strongly heterogeneous reservoirs, providing a reliable theoretical and technical basis for the efficient development and remaining oil recovery in such reservoirs during the late production stage. Full article

Traditional methods for predicting surrounding rock grades ahead of tunnel faces encounter challenges: image-based approaches are susceptible to environmental interference, while parameter-based classification may disrupt construction. This study proposes an intelligent rock grade identification method by integrating longitudinal displacement profile (LDP) evolution patterns with deep learning. First, the numerical model was validated against V-D theoretical curves, and LDP evolution laws were systematically analyzed for three rock types (GSI = 15, 30, 50) under nine geological combinations. The results indicate that (1) homogeneous strata exhibit deformation peaks followed by declines; (2) GSI = 15 strata show significantly larger deformations; and (3) stratified schemes display pre-interface deformation peaks and post-interface deformation controlled by subsequent lithology. A novel hybrid neural network was developed to classify strata using LDP curves as input. The model achieved 93.25% training accuracy and 91.20% validation accuracy. Ablation experiments demonstrated their superiority over the other four models with partial module deletions, achieving improvements in test accuracy of 3.24%, 3.08%, 4.16%, and 6.48%, respectively, compared to those models. This lightweight solution effectively overcomes the limitations of manual expertise dependency in conventional models and environmental sensitivity in visual methods. By synergizing LDP evolution analysis with deep learning, this framework provides a reliable approach for real-time rock grade prediction during tunnel advancement. Full article

Due to the fracture caverns in fractured-vuggy reservoirs, channeling frequently occurs during water injection or gas injection. The stability of the oil–water/oil–gas interface during water injection or gas injection in fractured-vuggy reservoirs significantly affects the displacement efficiency. However, there is a lack of in-depth understanding of the stability of the gas–water interface migration during water injection or gas injection in such reservoirs. In order to deal with this problem, this study combines indoor 3D visualization physical simulation experiment and fracture-cavity reservoir flow simulation. The law of interface transport and oil recovery in the process of injection of gas/water considering the degree of filling of fracture holes was studied and the influence of formation on crude oil viscosity, gas injection speed, inclination angle and other factors on the stability of the interface was compared. Results show that, under the influence of gravity differentiation, the oil–water interface of high-viscosity crude oil fluctuates obviously after water breakthrough, and the oil–water interface tends to be unstable, forming uneven oil cones. By reducing the gas drive speed, water invasion can be effectively inhibited to achieve a stable interface which accordingly improves the oil recovery. Full article

An electroencephalogram (EEG), recorded on the surface of the scalp, serves to characterize the distribution of electric potential during brain activity. This method finds extensive application in investigating brain functioning and diagnosing various diseases. Event-related potential (ERP) is employed to delineate visual, motor, and other activities through cross-trial averages. Despite its utility, interpreting the spatiotemporal dynamics in EEG data poses challenges, as they are inherently subject-specific and highly variable, particularly at the level of individual trials. Conventionally associated with oscillating brain sources, these dynamics raise questions regarding how these oscillations give rise to the observed dynamical regimes on the brain surface. In this study, we propose a model for spatiotemporal dynamics in EEG data using the Poisson equation, with the right-hand side corresponding to the oscillating brain sources. Through our analysis, we identify primary dynamical regimes based on factors such as the number of sources, their frequencies, and phases. Our numerical simulations, conducted in both 2D and 3D, revealed the presence of standing waves, rotating patterns, and symmetric regimes, mirroring observations in EEG data recorded during picture naming experiments. Notably, moving waves, indicative of spatial displacement in the potential distribution, manifested in the vicinity of brain sources, as was evident in both the simulations and experimental data. In summary, our findings support the conclusion that the brain source model aptly describes the spatiotemporal dynamics observed in EEG data. Full article

In factories, human posture recognition facilitates human–machine collaboration, human risk management, and workflow improvement. Compared to optical sensors, inertial sensors have the advantages of portability and resistance to obstruction, making them suitable for factories. However, existing product-level inertial sensing solutions are generally expensive. This paper proposes a low-cost human motion capture system based on BMI 160, a type of six-axis inertial measurement unit (IMU). Based on WIFI communication, the collected data are processed to obtain the displacement of human joints’ rotation angles around XYZ directions and the displacement in XYZ directions, then the human skeleton hierarchical relationship was combined to calculate the real-time human posture. Furthermore, the digital human model was been established on Unity3D to synchronously visualize and present human movements. We simulated assembly operations in a virtual reality environment for human posture data collection and posture recognition experiments. Six inertial sensors were placed on the chest, waist, knee joints, and ankle joints of both legs. There were 16,067 labeled samples obtained for posture recognition model training, and the accumulated displacement and the rotation angle of six joints in the three directions were used as input features. The bi-directional long short-term memory (BiLSTM) model was used to identify seven common operation postures: standing, slightly bending, deep bending, half-squatting, squatting, sitting, and supine, with an average accuracy of 98.24%. According to the experiment result, the proposed method could be used to develop a low-cost and effective solution to human posture recognition for factory operation. Full article

This study proposes a tightly coupled multi-sensor Simultaneous Localization and Mapping (SLAM) framework that integrates RGB-D and inertial measurements to achieve highly accurate 6 degree of freedom (6DOF) metric localization in a variety of environments. Through the consideration of geometric consistency, inertial measurement unit constraints, and visual re-projection errors, we present visual-inertial-depth odometry (called VIDO), an efficient state estimation back-end, to minimise the cascading losses of all factors. Existing visual-inertial odometers rely on visual feature-based constraints to eliminate the translational displacement and angular drift produced by Inertial Measurement Unit (IMU) noise. To mitigate these constraints, we introduce the iterative closest point error of adjacent frames and update the state vectors of observed frames through the minimisation of the estimation errors of all sensors. Moreover, the closed-loop module allows for further optimization of the global attitude map to correct the long-term drift. For experiments, we collect an RGBD-inertial data set for a comprehensive evaluation of VID-SLAM. The data set contains RGB-D image pairs, IMU measurements, and two types of ground truth data. The experimental results show that VID-SLAM achieves state-of-the-art positioning accuracy and outperforms mainstream vSLAM solutions, including ElasticFusion, ORB-SLAM2, and VINS-Mono. Full article