Search Results (34)

Laser powder bed fusion (LPBF) is a widely utilized metal additive manufacturing (AM) process for fabricating intricate geometries with high mechanical strength. However, achieving defect-free parts remains challenging due to complex thermodynamics and process variability. Component quality is primarily determined by mel-pool morphology, which depends on key process parameters such as laser power, scan speed, and layer thickness. Improper parameter selection causes defects like porosity (keyhole and lack of fusion), balling, and residual stresses, compromising structural integrity. Optimizing these parameters is crucial but difficult due to the multi-scale, multi-physics nature of the process, which traditionally relies on costly, time-intensive experimental trials. We present results from a data-driven approach using machine learning (ML) models to predict and optimize LPBF melt-pool characteristics, reducing reliance on trial-and-error experimentation. We find that laser power and scan speed predominantly influence the melt-pool formation. Higher scan speeds produce more favorable melt pools, whereas excessive laser power at low scan speeds leads to deep keyhole defects. To predict and classify melt pools efficiently, several ML models are deployed, including logistic regression, decision trees, ensemble learning, and fully connected neural networks. The standard neural network achieved the highest cross-validated macro-F1 score of 0.978 ± 0.014, while the weighted neural network achieved the highest recall for the rare optimal melt-pool class, 0.967 ± 0.050. These findings show that class-weighted learning provides a recall-oriented strategy for identifying suitable LPBF process windows, while avoiding overreliance on single train-test split performance. The findings underscore the effectiveness of ML in accurately classifying LPBF melt pools to rapidly identify optimal process parameters. Full article

Sugarcane harvester performance varies substantially with field geometry, crop, and operator factors, yet separating these sources from telematics data while preserving engineering interpretability remains a methodological gap. This study models field efficiency (Eff) and harvesting capacity (C_a) separately from JDLink telematics, aligning model structure with each target’s response behavior. Operational data covered 105 plots across four seasons (2019/20–2022/23) from three John Deere CH570 chopper harvesters in eastern Thailand. Six engineering-relevant predictors were retained after multicollinearity screening, and linear (MLR), additive nonlinear (GAM), and tree-based models were compared under 5-fold grouped cross-validation by BaseField (87 groups). Eff was assigned to GAM (R²_CV = 0.621 ± 0.114) on the basis of its threshold-like response to turning frequency; C_a was retained for MLR (R²_CV = 0.681 ± 0.121), with GAM essentially tied. Train–validation gaps were substantially smaller for additive models (0.096–0.118) than for tuned tree-based candidates (GBR 0.210–0.302, RF 0.322–0.358). Turning frequency (TF) and perimeter-to-area ratio (PAR) were the strongest predictors, and a constant-turn-time partial-out test indicated that TF’s univariate effect on Eff is largely mediated by the time-budget identity. Tactical interventions (path planning, operator training, machine–field allocation) are immediately feasible, although strategic field-layout change remains constrained by smallholder land tenure. Full article

Nature provides efficient strategies for fluid transport and thermal regulation through evolved structural features. This review summarizes recent progress in biomimetic thermal–fluid structures for enhancing fluid flow and heat transfer, with emphasis on the links among biological inspiration, engineering geometry, transport mechanisms, and application performance. Representative designs are classified into tree-like branching and fractal networks, compact hexagonal layouts, and bio-inspired curved morphologies, including riblets, grooves, fins, fluctuating channels, and TPMS structures. Their enhancement mechanisms involve flow redistribution, boundary-layer disturbance, secondary-flow and vortex generation, local acceleration, enlarged heat-transfer area, drag reduction, and compact flow organization. Applications using biomimetic structures are assessed in detail, such as in battery thermal management, electronic cooling, etc. The reviewed studies indicate that biomimetic structures can improve temperature uniformity, suppress hotspots, and enhance thermohydraulic performance, but the gains may be accompanied by pressure-drop or pumping-power penalties. Therefore, coupled thermal–hydraulic evaluation is essential for objective comparison. Key challenges of practical usage are identified in mechanism-based design, manufacturability, reliability, etc. This work establishes the guidance for translating biological forms into practical thermal–fluid structures with balanced efficacy. Full article

In advanced manufacturing processes, precise deposition behavior prediction is crucial for process parameter optimization. In order to forecast significant deposition responses such as bead width (w), bead height (h), energy input (EI), and volumetric input (VI) based on process parameters like laser power (P), travel speed (v), and wire feed rate (fw), seven different machine learning (ML) models were developed in this study, including Random Forest (RF), Gradient Boosting (GB), Extreme Gradient Boosting (XGB), LightGBM (LGBM), Extra Trees (ET), Support Vector Regression (SVR), and Elastic Net (EN). The predictive power of all models was assessed using the mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R²). The discoveries showed that ensemble models performed better than traditional ML techniques. The GB model performed the best overall, followed by the XGB model, which showed strong generalization and high prediction accuracy across training, validation, and testing datasets. Additionally, computational efficiency research discovered that the GB model holds a moderate model size and practically quick training time, making it suitable for real-world application. The robustness of the chosen model was supported by a paired t-test, which verified that the performance differences between the GB model and other models are statistically significant (p < 0.05). Furthermore, the impact of input parameters on the anticipated responses was interpreted using SHAP (SHapley Additive exPlanations) analysis. The interpretation results exhibited that while wire feed rate mostly affects volumetric deposition behavior, laser power and travel speed are the key parameters monitoring energy input and bead geometry. The GB and XGB models were used to forecast deposition reactions using specific process parameters, and the predictions were compared with experimental findings in order to further confirm the predictive power of the developed models with better performance. Overall, the results display that this study, in combination with ensemble boosting models, offers a reliable framework for understanding complicated relationships between deposition features and processing parameters, providing insightful information for process optimization in advanced manufacturing applications. Full article

The semantic segmentation of sparse array synthetic aperture radar (SAR) 3D point clouds remains a significant challenge. These datasets are characterized by extreme sparsity, irregular distribution, and structural discontinuity, factors that diminish the reliability of local neighborhoods and impede the performance of traditional segmentation algorithms. This study introduces an enhanced PointNet++ framework specifically tailored for the semantic segmentation of sparse array-SAR 3D point clouds. Utilizing PointNet++ as a hierarchical backbone, the proposed architecture incorporates three geometry-oriented modifications: a feature enhancement strategy integrating normalized height, surface normals, and local density; an EdgeConv module positioned at an intermediate abstraction stage to reinforce local geometric modeling; and an FP-Refine module designed to optimize cross-scale feature propagation and recovery within sparse regions. Rather than proposing a fundamentally distinct universal architecture, this research focuses on a task-oriented adaptation of PointNet++ to address the neighborhood instability and structural gaps inherent in sparse array-SAR data. Experimental evaluations using the SARMV3D-1.0 dataset indicate that the proposed method consistently outperforms the PointNet++ baseline, maintaining stable performance across various random seeds with an mIoU between 55% and 58%. Further validation through ablation studies, parameter sensitivity analyses, and perturbation-based robustness assessments confirms the utility of the integrated components. Additionally, cross-dataset experiments on S3DIS and Toronto3D suggest that the framework generalizes effectively to point clouds with varying densities and spatial configurations. The findings demonstrate that the method is particularly successful for categories defined by distinct vertical geometry and structural continuity, such as trees, roofs, and facades, though performance remains limited for weakly structured classes like roads. Full article

Aerial remote sensing efficiently surveys large areas, but accurate direct object-level measurement remains difficult in complex natural scenes. Advancements in 3D computer vision, particularly radiance field representations such as NeRF and 3D Gaussian splatting, can improve reconstruction fidelity from posed imagery. Nevertheless, direct aerial measurement of important attributes like tree diameter at breast height (DBH) remains challenging. Trunks in aerial forest scans are distant and sparsely observed in image views; at typical operating altitudes, stems may span only a few pixels. With these constraints, conventional reconstruction methods have inaccurate breast-height trunk geometry. TreeDGS is an aerial image reconstruction method that uses 3D Gaussian splatting as a continuous scene representation for trunk measurement. After SfM–MVS initialization and Gaussian optimization, we extract a dense point set from the Gaussian field using RaDe-GS’s depth-aware cumulative-opacity integration and associate each sample with a multi-view opacity reliability score. Then, we isolate trunk points and estimate DBH using opacity-weighted solid-circle fitting. Evaluated on 10 plots with field-measured DBH, TreeDGS reaches 4.79 cm RMSE (about 2.6 pixels at this GSD) and outperforms a LiDAR baseline (7.66 cm RMSE). This shows that TreeDGS can enable accurate, low-cost aerial DBH measurement. Full article

Mapping individual-tree crowns (ITCs) along with extracting tree morphological attributes provides the core parameters required for estimating thermal stress and carbon emission functions. However, calculating morphological attributes relies on the prior delineation of ITCs. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) framework, this review synthesizes how deep-learning (DL)-based methods enable the conversion of crown geometry into reliable biometric parameter extraction (BPE) from high-resolution imagery. This addresses a gap often overlooked in studies focused solely on detection by providing a direct link to forest inventory metrics. Our review showed that instance segmentation dominates (approximately 46% of studies), producing the most accurate pixel-level masks for BPE, while RGB imagery is most common (73%), often integrated with canopy-height models (CHM) to enhance accuracy. New architectural approaches, such as StarDist, outperform Mask R-CNN by 6% in dense canopies. However, performance differs with crown overlap, occlusion, species diversity, and the poor transferability of allometric equations. Future work could prioritize multisensor data fusion, develop end-to-end biomass modeling to minimize allometric dependence, develop open datasets to address model generalizability, and enhance and test models like StarDist for higher accuracy in dense forests. Full article

The development of shale gas reservoirs relies on complex fracture networks created via multistage hydraulic fracturing, yet most existing models still use oversimplified fracture geometries and therefore cannot fully capture the coupled effects of multiscale fracture topology on flow and production. To address this gap, in this study, we combine fractal geometry with the Embedded Discrete Fracture Model (EDFM) to analyze the production dynamics of hydraulically fractured horizontal wells in shale gas reservoirs. A tree-like fractal fracture network is first generated using a stochastic fractal growth algorithm, where the iteration number, branching number, scale factor, and deviation angle control the self-similar hierarchical structure and spatial distribution of fractures. The resulting fracture network is then embedded into an EDFM-based, fully implicit finite-volume simulator with Non-Neighboring Connections (NNCs) to represent multiscale fracture–matrix flow. A synthetic shale gas reservoir model, constructed using representative geological and engineering parameters and calibrated against field production data, is used for all numerical experiments. The results show that increasing the initial water saturation from 0.20 to 0.35 leads to a 26.4% reduction in cumulative gas production due to enhanced water trapping. Optimizing hydraulic fracture spacing to 200 m increases cumulative production by 3.71% compared with a 100 m spacing, while longer fracture half-lengths significantly improve both early-time and stabilized gas rates. Increasing the fractal iteration number from 1 to 3 yields a 36.4% increase in cumulative production and markedly enlarges the pressure disturbance region. The proposed fractal–EDFM framework provides a synthetic yet field-calibrated tool for quantifying the impact of fracture complexity and design parameters on shale gas well productivity and for guiding fracture network optimization. Full article

Tree-like fractals as internal structures are a novel alternative to conventional lattice structures for mechanical components produced via Selective Laser Melting (SLM). This study explores the mechanical behavior of tree-like fractals, targeting flexure tests on SLM test samples manufactured using two distinct fractal configurations. The main objective is to develop numerical models that can predict the effect of the branching angle on the stress-strain curves, for both fractal configurations, from experimental flexure tests. A polynomial regression model is proposed to predict mechanical response variations based on fractal geometry, and the prediction model provides acceptable errors, less than the natural variance of multiple experiments. Furthermore, the tree-like fractal samples showed an interesting behavior on the flexure test, where the fractals deformed uniformly and in a predictable pattern, enabling mechanical advantages in impact absorption applications. Full article

Reinforced concrete (RC) tanks are essential for storing liquids and bulk materials across various industries. However, simplified analytical methods fall short in providing an accurate analysis, while traditional methods, such as finite element modeling, can be computationally intensive and time-consuming, especially when dealing with nonlinear material properties and complex geometries, like conical and cylindrical shapes. This highlights the need for a more efficient and simplified analysis approach. Accordingly, the present paper introduces a machine learning (ML) framework as an effective predictive tool for RC conical and cylindrical tanks under hydrostatic pressure. Data from 320 RC conical and cylindrical water tanks, previously analyzed using finite element modeling, were used to train and test various ML models, considering geometrical and material nonlinearities. Four machine learning models—decision trees, random forests, gradient boosting, and extreme gradient boosting—were utilized to predict critical internal forces, including the maximum ring tension force, maximum meridional moment, and maximum meridional axial force. The accuracy of each model was evaluated using different statistical measures. To improve model interpretability and identify key predictors, feature importance techniques were employed to rank the significance of each input variable to the predictions. Furthermore, Accumulated Local Effects (ALE) plots were utilized to visualize the relationships between model inputs and outputs, providing a clearer understanding of the inner workings of the ML models. The combined use of feature importance and ALE plots enhances model transparency by illustrating how specific features contribute to the predictions, thereby supporting the informed application of ML in the structural design and analysis of RC tanks. Ultimately, the framework presented in this study aims to promote the practical application of machine learning in structural engineering, contributing to simpler, more efficient, and accurate analysis and design processes for RC water tanks. Full article

Monocular depth estimation (MDE) is a critical computer vision task that enhances environmental perception in fields such as autonomous driving and robot navigation. In recent years, deep learning-based MDE methods have achieved notable progress in these fields. However, achieving robust monocular depth estimation in low-altitude forest environments remains challenging, particularly in scenes with dense and cluttered foliage, which complicates applications in environmental monitoring, agriculture, and search and rescue operations. This paper presents a comprehensive evaluation of state-of-the-art deep learning-based MDE methods on low-altitude forest datasets. The evaluated models include both self-supervised and supervised approaches, employing different network structures such as convolutional neural networks (CNNs) and Vision Transformers (ViTs). We assessed the generalization of these approaches across diverse low-altitude scenarios, specifically focusing on forested environments. A systematic set of evaluation criteria is employed, comprising traditional image-based global statistical metrics as well as geometry-aware metrics, to provide a more comprehensive evaluation of depth estimation performance. The results indicate that most Transformer-based models, such as DepthAnything and Metric3D, outperform traditional CNN-based models in complex forest environments by capturing detailed tree structures and depth discontinuities. Conversely, CNN-based models like MiDas and Adabins struggle with handling depth discontinuities and complex occlusions, yielding less detailed predictions. On the Mid-Air dataset, the Transformer-based DepthAnything demonstrates a 54.2% improvement in RMSE for the global error metric compared to the CNN-based Adabins. On the LOBDM dataset, the CNN-based MiDas has the depth edge completeness error of 93.361, while the Transformer-based Metric3D demonstrates the significantly lower error of only 5.494. These findings highlight the potential of Transformer-based approaches for monocular depth estimation in low-altitude forest environments, with implications for high-throughput plant phenotyping, environmental monitoring, and other forest-specific applications. Full article

A novel sandwich structure of a fractal tree-like lattice (SSFL) is proposed. The geometry characteristics were constructed based on the fractal tree-like patterns found in many biological structures, such as giant water lilies and dragon blood trees. The compressive performance of the proposed structures with different fractal orders was experimentally and numerically investigated. The experimental samples were made by 3D printing technology. Axial compression tests were conducted to study the compressive performance and failure mode of the SSFLs. The results indicated that the new structure was good at multiple bearing and energy absorption. The finite element method (FEM) was performed to investigate the influence of geometry parameters on the compression behaviors of the SSFLs. The findings of this study provide an effective guide for using the fractal method to design lattice structures with a high bearing capacity. Full article

Three-dimensional laser scanning technology has increasingly gained favor among professionals in tunnel monitoring. A fundamental challenge in tunnel point cloud processing is to efficiently manage massive datasets using appropriate data structures and accurately extract features such as tunnel axes and cross-sections. However, existing studies often disconnect tunnel point cloud indexing from post-processing tasks. Conventional structures (e.g., voxels, octrees) struggle with long strip-like uneven spatial distribution, resulting in imbalanced trees with numerous empty nodes, which are incompatible with axis-aligned operations. Therefore, this study proposes a dual-layer indexing structure tailored to tunnel geometries. The upper layer reorganizes the tunnel point cloud along its axis, while the lower layer leverages local octrees for fast data querying and updates. In implementation, we introduce a merge-based octree generation strategy for ultra-large-scale datasets, and a rapid Hough transform-based algorithm for tunnel boundaries and axes extraction. Experimental results demonstrate that the proposed method successfully supports the management and visualization of a tunnel point cloud exceeding 6 billion points, significantly enhancing efficiency in narrow tunnel scenarios and streamlining various axis-aligned post-processing tasks. Full article

The Delottococcus aberiae is a mealybug pest known as Cotonet de les Valls in the province of Castellón (Spain). This tiny insect is causing large economic losses in the Spanish agricultural sector, especially in the citrus industry. The European Copernicus program encourages the progress of Earth observation (EO) in relation to the development of agricultural monitoring tools. In this context, this work is based on the analysis of the temporal evolution of spectral surface reflectance data from Sen2Like, analyzing healthy and fields affected by the mealybug. The study area is focused on the surroundings of Vall d’Uixó (Castellón, Spain), involving an approximate area of 25 ha distributed in a total of 21 fields of citrus trees with different mealybug incidence, classified as healthy or unhealthy, during the 2020–2021 season. The relationship between the mealybug infestation level and the Normalized Difference Vegetation Index (NDVI) and other optical bands (Red, NIR, SWIR, derived from Sen2Like) were analyzed by studying the time-series evolution of each parameter across the time period 2017–2022. In this study, we also demonstrate that evergreen fruit trees such as citrus, show a seasonality across the EO-based time series, which is linked to directional effects caused by the sensor–sun geometry. This can be mitigated by using a Bidirectional Reflectance Distribution Function (BRDF) model such as the High-Resolution Adjusted BRDF Algorithm (HABA). To study the infested fields separately from healthy ones and avoid mixing fields with very different spectral responses caused by field type, separation between rows, or age, we studied the evolution of each parcel separately using monthly linear regressions, considering the 2017–2018 seasons as a reference when the pest had not developed yet. The observations indicate the feasibility of the distinction between affected and healthy plots during a year utilizing specific spectral ranges, with SWIR proving a notably effective channel, enabling separability from mid-summer to the fall. Furthermore, the anomaly inspection demonstrates an increase in the effects of the pest from 2020 to 2022 in all spectral regions and enables a first approximation for identifying healthy and affected fields based on negative anomalies in the red and SWIR channels and positive anomalies in the NIR and NDVI. This work contributes to the development of new monitoring tools for efficient and sustainable action in pest control. Full article

Geometrical models of the airways offer a comprehensive perspective on the complex interplay between lung structure and function. Originating from mathematical frameworks, these models have evolved to include detailed lung imagery, a crucial enhancement that aids in the early detection of morphological changes in the airways, which are often the first indicators of diseases. The accurate representation of airway geometry is crucial in research areas such as biomechanical modeling, acoustics, and particle deposition prediction. This review chronicles the evolution of these models, from their inception in the 1960s based on ideal mathematical constructs, to the introduction of advanced imaging techniques like computerized tomography (CT) and, to a lesser degree, magnetic resonance imaging (MRI). The advent of these techniques, coupled with the surge in data processing capabilities, has revolutionized the anatomical modeling of the bronchial tree. The limitations and challenges in both mathematical and image-based modeling are discussed, along with their applications. The foundation of image-based modeling is discussed, and recent segmentation strategies from CT and MRI scans and their clinical implications are also examined. By providing a chronological review of these models, this work offers insights into the evolution and potential future of airway geometry modeling, setting the stage for advancements in diagnosing and treating lung diseases. This review offers a novel perspective by highlighting how advancements in imaging techniques and data processing capabilities have significantly enhanced the accuracy and applicability of airway geometry models in both clinical and research settings. These advancements provide unique opportunities for developing patient-specific models. Full article