Search Results (14,422)

Real-time object detection in Unmanned Aerial Vehicle (UAV) imagery is critical yet challenging, requiring high accuracy amidst complex scenes with multi-scale and small objects, under stringent onboard computational constraints. While existing methods struggle to balance accuracy and efficiency, we propose RTUAV-YOLO, a family of lightweight models based on YOLOv11 tailored for UAV real-time object detection. First, to mitigate the feature imbalance and progressive information degradation of small objects in current architectures multi-scale processing, we developed a Multi-Scale Feature Adaptive Modulation module (MSFAM) that enhances small-target feature extraction capabilities through adaptive weight generation mechanisms and dual-pathway heterogeneous feature aggregation. Second, to overcome the limitations in contextual information acquisition exhibited by current architectures in complex scene analysis, we propose a Progressive Dilated Separable Convolution Module (PDSCM) that achieves effective aggregation of multi-scale target contextual information through continuous receptive field expansion. Third, to preserve fine-grained spatial information of small objects during feature map downsampling operations, we engineered a Lightweight DownSampling Module (LDSM) to replace the traditional convolutional module. Finally, to rectify the insensitivity of current Intersection over Union (IoU) metrics toward small objects, we introduce the Minimum Point Distance Wise IoU (MPDWIoU) loss function, which enhances small-target localization precision through the integration of distance-aware penalty terms and adaptive weighting mechanisms. Comprehensive experiments on the VisDrone2019 dataset show that RTUAV-YOLO achieves an average improvement of 3.4% and 2.4% in mAP50 and mAP50-95, respectively, compared to the baseline model, while reducing the number of parameters by 65.3%. Its generalization capability for UAV object detection is further validated on the UAVDT and UAVVaste datasets. The proposed model is deployed on a typical airborne platform, Jetson Orin Nano, providing an effective solution for real-time object detection scenarios in actual UAVs. Full article

Reliable water access in remote and desert-like regions remains a challenge, particularly in areas with limited infrastructure. Solar-powered submersible pumps offer a promising solution; however, optimizing their performance under variable environmental conditions remains a challenging task. This research presents an Artificial Intelligence (AI)-driven digital twin framework for modeling and optimizing the performance of a solar-powered submersible pump system. The proposed system has three core components: (1) an AI model for predicting the inverter motor’s output frequency based on the current generated by the solar panels, (2) a predictive model for estimating the pump’s generated power based on the inverter motor’s output, and (3) a mathematical formulation for determining the volume of water lifted based on the system’s operational parameters. Moreover, a dataset comprising 6 months of environmental and system performance data was collected and utilized to train and evaluate multiple predictive models. Unlike previous works, this research integrates real-world data with a multi-phase AI modeling pipeline for real-time water output estimation. Performance assessments indicate that the Random Forest (RF) model outperformed alternative approaches, achieving the lowest error rates with a Mean Absolute Error (MAE) of 1.00 Hz for output frequency prediction and 1.39 kW for pump output power prediction. The framework successfully estimated annual water delivery of 166,132.77 m³, with peak monthly output of 18,276.96 m³ in July and minimum of 9784.20 m³ in January demonstrating practical applicability for agricultural water management planning in arid regions. Full article

The geometric content of chaos in nonlinear systems with multiple stabilities of high order is a challenge to computation. We introduce a single algorithmic framework to overcome this difficulty in the present study, where a parametrically forced oscillator with cubic–quintic nonlinearities is considered as an example. The framework starts with the Sparse Identification of Nonlinear Dynamics (SINDy) algorithm, which is a self-learned algorithm that extracts an interpretable and correct model by simply analyzing time-series data. The resulting parsimonious model is well-validated, and besides being highly predictive, it also offers a solid base on which one can conduct further investigations. Based on this tested paradigm, we propose a unified diagnostic pathway that includes bifurcation analysis, computation of the Lyapunov exponent, power spectral analysis, and recurrence mapping to formally describe the dynamical features of the system. The main characteristic of the framework is an effective algorithm of computational basin analysis, which is able to display attractor basins and expose the fine scale riddled structures and fractal structures that are the indicators of extreme sensitivity to initial conditions. The primary contribution of this work is a comprehensive dynamical analysis of the DM-CQDO, revealing the intricate structure of its stability landscape and multi-stability. This integrated workflow identifies the period-doubling cascade as the primary route to chaos and quantifies the stabilizing effects of key system parameters. This study demonstrates a systematic methodology for applying a combination of data-driven discovery and classical analysis to investigate the complex dynamics of parametrically forced, high-order nonlinear systems. Full article

As high-energy-consumption underground public space, subway stations are responsible for a particularly significant proportion of air-conditioning energy use, especially during the cooling season, making the investigation of cooling-load characteristics highly important. However, the determination of independent influencing factors in different situations has not yet reached a consensus, and the role of interaction effects is lacking, which hinders the development of energy-saving strategies. For this purpose, this study proposes a sensitivity analysis framework based on 10 typical influencing factors from thermal parameters, meteorological parameters, internal heat disturbances, and indoor environmental setpoints. An input set was generated by integrating equal-step parameter discretization and Saltelli quasi-MonteCarlo sampling. A database containing 11,264 samples was constructed through an EnergyPlus–Python co-simulation method. Based on the Sobol global sensitivity analysis, the key influencing factors of subway station cooling load were identified and quantified, and the impact of these 10 factors was systematically analyzed. Results show that occupant density (S_i^T = 0.5605) and fresh air volume (S_i^T = 0.4546) are the dominant factors, contributing more than 50% of the load variance. In contrast, the characteristics of an underground structure significantly weaken the influence of the building-envelope heat transfer coefficient (S_i^T = 0.1482) and soil temperature (S_i^T = 0.0884). Furthermore, five groups of strong interaction effects were identified in this study, including occupant density–fresh air volume (S_ij = 0.1094), revealing a nonlinear load response mechanism driven by multi-parameter coupling. This research provides a theoretical foundation and quantitative tool for the refined design and optimized dynamic coupled operation of underground transportation hubs. Full article

The line contact behavior of multiscale meshing interfaces necessitates synergistic investigation spanning nano-to centimeter-scale ranges. When nominally smooth gear teeth surfaces come into contact, the mechanical–thermal coupling effect at the meshing interface actually occurs over a collection of microscale asperities (roughness peaks) exhibiting hierarchical distribution characteristics. The emergent deformation phenomena across multiple asperity scales govern the self-organized evolution of interface conformity, thereby regulating both the load transfer efficiency and thermal transport properties within the contact zone. The fractal nature of the roughness topography on actual meshing interfaces calls for the development of a cross-scale theoretical framework that integrates micro-texture optimization with multi-physics coupling contact behavior. Conventional roughness characterization methods based on statistical parameters suffer from inherent limitations: their parameter values are highly dependent on measurement scale, lacking uniqueness under varying sampling intervals and instrument resolutions, and failing to capture the scale-invariant nature of meshing interface topography. A scale-independent parameter system grounded in fractal geometry theory enables essential feature extraction and quantitative characterization of three-dimensional interface morphology. This study establishes a progressive deformation theory for gear line contact interfaces with fractal geometric characteristics, encompassing elastic, elastoplastic transition, and perfectly plastic stages. By systematically investigating the force–thermal coupling mechanisms in textured meshing interfaces under multiscale conditions, the research provides a theoretical foundation and numerical implementation pathways for high-precision multiscale thermo-mechanical analysis of meshing interfaces. Full article

This paper introduces TRIDENT-DE, a novel ensemble-based variant of Differential Evolution (DE) designed to tackle complex continuous global optimization problems. The algorithm leverages three complementary trial vector generation strategies best/1/bin, current-to-best/1/bin, and pbest/1/bin executed within a self-adaptive framework that employs jDE parameter control. To prevent stagnation and premature convergence, TRIDENT-DE incorporates adaptive micro-restart mechanisms, which periodically reinitialize a fraction of the population around the elite solution using Gaussian perturbations, thereby sustaining exploration even in rugged landscapes. Additionally, the algorithm integrates a greedy line-refinement operator that accelerates convergence by projecting candidate solutions along promising base-to-trial directions. These mechanisms are coordinated within a mini-batch update scheme, enabling aggressive iteration cycles while preserving diversity in the population. Experimental results across a diverse set of benchmark problems, including molecular potential energy surfaces and engineering design tasks, show that TRIDENT-DE consistently outperforms or matches state-of-the-art optimizers in terms of both best-found and mean performance. The findings highlight the potential of multi-operator, restart-aware DE frameworks as a powerful approach to advancing the state of the art in global optimization. Full article

The paper studies the dynamic characteristics of a multi-layer foil thrust bearing (MLFTB). A modified efficient dynamic characteristic model is established, and the revised Reynolds equation coupled with the thick plate element and the boundary slip model is adopted. During the solving process, the small perturbation method is implemented. The elasto-hydrodynamic effect under geometric and operational parameters is investigated. It reflects that the dynamic characteristics can be visibly influenced by the slip effect when under tiny clearance with low bearing speed, and ought to be considered. Specifically, the maximum deviation of the axial and direct-rotational stiffness coefficients could be up to −4.93% and −5.02%, respectively. The direct-rotational stiffness is increased with the perturbation frequency; however, a turning point may exist in the cross-rotational stiffness. Additionally, both the rotational stiffness and rotational damping can be expanded at a smaller original clearance. It aims to provide prediction methods with high effectiveness and efficiency, and enrich theoretical guidance for the important MLFTB. Full article

Efficient acquisition of rock mass parameters is a critical step for conducting numerical simulations in tunneling and ensuring the safety of subsequent construction. This paper proposes an intelligent back-analysis method for key rock mass parameters (Young’s modulus, Poisson’s ratio, cohesion, and friction angle) based on excavation-induced deformation data, using a deformation database that incorporates multi-feature values from tunnel excavation. This study employs five machine learning algorithms with single-feature inputs and three deep neural networks (DNNs) with multi-feature inputs, with a particular focus on convolutional neural network (CNN) due to their superior performance in terms of both accuracy and efficiency. The results demonstrate that the CNN model incorporating excavation features achieves excellent performance in parameter back-analysis, with an R² of 0.99 and 97.8% of predictions having errors within 5%. Compared with machine learning models using single-feature inputs, the CNN-based approach improves predictive performance by an average of 13.9%. Furthermore, compared with other DNNs, the CNN consistently outperforms across various evaluation metrics. This study also investigates the CNN’s capability to predict rock mass parameters using deformation data from early-stage excavation. After ten excavation steps, 96.9% of test samples had prediction errors within 5%. Finally, the proposed method was validated using field-monitored deformation data from a real highway tunnel project, confirming the method’s effectiveness and practical applicability. Full article

The expanding use of rare earth elements (REEs) in high-tech industrials has increased their environmental release, raising concerns about their ecological risks. This study employed the Diffusive Gradients in Thin Films (DGT) technique to assess REE bioavailability, spatial distribution, and ecological risks of REEs in sediments of the Yitong River, a historically polluted urban river in Changchun, China. Sediment characteristics (organic matter, pH, salinity), nutrient dynamics (N, P), and metal concentrations (Fe, Mn, As, etc.) were analyzed alongside REEs to evaluate their interactions and environmental drivers. Results revealed that REE concentrations (0.453–1.687 μg L⁻¹) were dominated by light REEs (50.1%), with levels an order of magnitude lower than heavily industrialized regions. Ecological risk quotients (RQ) for individual REEs were below thresholds (RQ < 1), indicating negligible immediate risks, though spatial trends suggested urban runoff influences. Probabilistic risk assessment integrating DGT data and species sensitivity distributions (SSD) estimated a low combined toxic probability (2.26%) for REEs and nutrients. Microbial community analysis revealed correlations between specific bacterial (e.g., Clostridium, Dechloromonas) and fungal genera (e.g., Pseudeurotium) with metals and REEs, highlighting microbial sensitivity to pollutant shifts. This study provides a multidimensional framework linking REE bioavailability, sediment geochemistry, and microbial ecology, offering insights for managing REE contamination in urban riverine systems. Full article

The global transition to renewable energy requires hybrid solutions that address variability while delivering tangible co-benefits and verifiable mitigation outcomes. This study evaluates a novel small hydropower–photovoltaic (SHP–PV) hybrid system in the Kyrgyz Republic that integrates flexible Bitcoin mining loads and waste-heat recovery for greenhouse heating. A techno-economic model was developed for a 10 MW configuration, allocating annual net generation of 57.34 GWh between grid export and on-site mining through a single decision parameter. Mitigation accounting applies a combined margin grid factor of 0.4–0.7 tCO₂/MWh for exported electricity and a diesel factor of 0.26–0.27 tCO₂/MWh_fuel for heat displacement, yielding Article 6–eligible reductions from both electricity and recovered heat. Waste-heat recovery from mining supplies ≈15 MWh_th/year to a 50 m² greenhouse, displacing diesel use and demonstrating visible sustainable development co-benefits. Economic analysis reproduces annual revenues of ≈$1.9 million, with a levelized cost of electricity of $48/MWh and an indicative IRR of ~6%, consistent with positive but modest returns under merchant operation and uplift potential under mixed allocations. This study concludes that componentized accounting—exported electricity credited under grid displacement and diesel displacement credited from recovered heat—ensures Article 6 integrity and positions SHP–PV hybrids as replicable, multi-service renewable models for Central Asia. Unlike prior hybrid studies that treat generation, economics, and mitigation separately, our framework integrates allocation (α), financial outcomes, and Article 6 carbon accounting within a unified structure, while explicitly modeling Bitcoin mining as an endogenous flexible load with thermal recovery—advancing methodological approaches for multi-service renewable systems in climate policy contexts. Full article

The lack of detailed design information in legacy hydropower plants creates challenges for modernising their ageing turbine components. This research advances a digitalisation approach which combines inverse design methodology (IDM) with multi-objective genetic algorithms (MOGA) and computational fluid dynamics (CFD) to digitally reconstruct and optimise the Bérchules Francis turbine runner and guide vane geometries using limited available legacy data, avoiding invasive techniques. A two-stage optimisation process was conducted. The first stage of runner blade optimisation achieved a 22.7% reduction in profile loss and a 16.8% decrease in secondary flow factor while raising minimum pressure from −877,325.5 Pa to −132,703.4 Pa. Guide vane optimisation during Stage 2 produced additional performance gains through a 9.3% reduction in profile loss and a 20% decrease in secondary flow factor and a minimum pressure increase to +247,452.1 Pa which represented an 183% improvement. The CFD validation results showed that the final turbine efficiency reached 93.7% while producing more power than the plant’s rated 942 kW. The sensitivity analysis revealed that leading edge loading at mid-span and normal chord proved to be the most significant design parameters affecting pressure loss and flow behaviour metrics. The research proves that legacy turbines can be digitally restored through hybrid optimisation and CFD workflows, which enables data-driven refurbishment design without needing complete component replacement. Full article

Background: Attention-deficit/hyperactivity disorder (ADHD) is a heterogeneous neurodevelopmental condition that currently relies on subjective clinical judgment for diagnosis, emphasizing the need for objective, clinically applicable tools. Methods: We applied machine learning techniques to parent reports, self-reports, and performance-based measures in a sample of 255 Taiwanese children and adolescents (108 ADHD and 147 controls; mean age = 11.85 years). Models were trained under a nested cross-validation framework to avoid performance overestimation. Results: Most models achieved high classification accuracy (AUCs ≈ 0.886–0.906), while convergent feature importance across models highlighted parent-rated social problems, executive dysfunction, and self-regulation traits as robust predictors. Additionally, ex-Gaussian parameters derived from reaction time distributions on the Continuous Performance Test (CPT) proved more informative than raw scores. Conclusions: These findings support the utility of integrating multi-informant ratings and task-based measures in interpretable ML models to enhance ADHD diagnosis in clinical practice. Full article

Deteriorating check dams pose significant threats to human safety and property, while impeding eco-environmental restoration in soil–water conservation systems in vulnerable watersheds like the Jiuyuangou Basin on China’s Loess Plateau. This study aimed to develop a comprehensive risk assessment framework for the check dam system in the Jiuyuangou Basin, China, to mitigate its threats to safety and eco-environmental restoration. A multi-index and multilevel risk evaluation system was established for check dam systems in the Jiuyuangou Basin, utilizing data gathering, hydrological statistics, numerical computation, and various methodologies. The index weights were determined via the fuzzy analytic hierarchy process with an integrated modeling framework for key parameters. Finally, the risk level of the check dam system in the Jiuyuangou Basin was assessed based on the comprehensive score. The results show that (1) nearly half of the check dams are at mild risk, approximately 25% are at moderate risk, and a few are basically safe. (2) Among various types of risk, the distribution of engineering risk is relatively uniform, environmental risk is generally high, loss risk is relatively concentrated, and management risk is particularly prominent. This research provides a scientific foundation for optimizing check dam governance, enhancing sediment control, and strengthening ecological service functions in vulnerable watersheds. Full article

Ovarian follicular fluid (FF) is a metabolically active and biomarker-rich medium that mirrors the oocyte microenvironment. Its analysis is increasingly recognized in infertility diagnostics and assisted reproductive technologies (ART) for assessing oocyte competence, understanding reproductive disorders, and guiding personalized treatment. However, FF’s high viscosity, complex composition, and biochemical variability challenge reproducibility in sample preparation and molecular profiling. Sonication-based extraction has emerged as an effective approach to address these issues. By exploiting acoustic cavitation, sonication improves protein solubilization, metabolite release, and lipid recovery, while reducing solvent use and processing time. This review synthesizes recent advances in sonication-assisted FF analysis across proteomics, metabolomics, and lipidomics, emphasizing parameter optimization, integration with advanced mass spectrometry workflows, and emerging applications in microfluidics, automation, and point-of-care devices. Clinical implications are discussed in the context of enhanced biomarker discovery pipelines, real-time oocyte selection, and ART outcome prediction. Key challenges, such as preventing biomolecule degradation, standardizing protocols, and achieving inter-laboratory reproducibility, are addressed alongside regulatory considerations. Future directions highlight the potential of combining sonication with multi-omics strategies and AI-driven analytics, paving the way for high-throughput, standardized, and clinically actionable FF analysis to advance precision reproductive medicine. Full article

Creating the most accurate numerical models with the same dynamic behavior as real structures plays an important role in the development process of various facilities. This article deals with the use of experimental methods, particularly experimental modal analysis (EMA), scanning, detection, spectral analysis, and mechanical testing in combination with the optimization techniques of the ANSYS 2024 R1 software to calibrate numerical models of axisymmetric structures. The proposed methodology was tested on a steel pipe whose geometric and material properties were both available. Within the updating of finite element models (FEMU) with one or two design variables, the influence of the range of feasible values on the accuracy of the observed parameters was examined. The updating process led to the acquisition of such a pipe model, which natural frequencies differed by less than 1.5% from the results estimated in EMA, and its weight also differed only minimally. The proposed methodology was then used for the FEMU of a pressure vessel whose contour was obtained by a 3D scanning method; material properties were investigated, and all wall thicknesses, i.e., eleven design variables, were unknown and thus determined by an iterative optimization technique. Using the Multi-Objective Genetic Algorithm (MOGA) method, the dimensions of the vessel were first updated for their shell model and subsequently for the 3D model. The resulting natural frequencies of the model with applied internal pressures of 0 bar, 40 bar, and 80 bar differed from those estimated experimentally by less than 1.2%. Full article