Search Results (9,412)

Recent developments in next-generation sequencing technology have led to the creation of extensive, open-source protein databases consisting of hundreds of millions of sequences. To render these sequences applicable in biomedical applications, they must be meticulously annotated by wet lab testing or extracting them from existing literature. Over the last few years, researchers have developed numerous automatic annotation systems, particularly deep learning models based on machine learning and artificial intelligence, to address this issue. In this work, we propose a transformer-based fusion model capable of predicting Gene Ontology (GO) terms from full-scale protein sequences, achieving state-of-the-art accuracy compared to other contemporary machine learning annotation systems. The approach performs particularly well on clustered split datasets, which comprise training and testing samples originating from distinct distributions that are structurally diverse. This demonstrates that the model is able to understand both short and long term dependencies within the protein’s structure and can capture sequence features that are predictive of the various GO terms. Furthermore, the technique is lightweight and less computationally expensive compared to the benchmark methods, while at the same time unaffected by sequence length, rendering it appropriate for diverse applications with varying sequence lengths. Full article

Accurate aircraft trajectory prediction is fundamental to air traffic management, operational safety, and intelligent aerospace systems. With the growing availability of flight data, deep learning has emerged as a powerful tool for modeling the spatiotemporal complexity of 4D trajectories. This paper presents a comprehensive review of deep learning-based approaches for aircraft trajectory prediction, focusing on their evolution, taxonomy, performance, and future directions. We classify existing models into five groups—RNN-based, attention-based, generative, graph-based, and hybrid and integrated models—and evaluate them using standardized metrics such as the RMSE, MAE, ADE, and FDE. Common datasets, including ADS-B and OpenSky, are summarized, along with the prevailing evaluation metrics. Beyond model comparison, we discuss real-world applications in anomaly detection, decision support, and real-time air traffic management, and highlight ongoing challenges such as data standardization, multimodal integration, uncertainty quantification, and self-supervised learning. This review provides a structured taxonomy and forward-looking perspectives, offering valuable insights for researchers and practitioners working to advance next-generation trajectory prediction technologies. Full article

Accurate real-time detection of blueberry maturity is vital for automated harvesting. However, existing methods often fail under occlusion, variable lighting, and dense fruit distribution, leading to reduced accuracy and efficiency. To address these challenges, we designed a lightweight deep learning framework that integrates improved feature extraction, attention-based fusion, and progressive transfer learning to enhance robustness and adaptability To overcome these challenges, we propose BMDNet-YOLO, a lightweight model based on an enhanced YOLOv8n. The backbone incorporates a FasterPW module with parallel convolution and point-wise weighting to improve feature extraction efficiency and robustness. A coordinate attention (CA) mechanism in the neck enhances spatial-channel feature selection, while adaptive weighted concatenation ensures efficient multi-scale fusion. The detection head employs a heterogeneous lightweight structure combining group and depthwise separable convolutions to minimize parameter redundancy and boost inference speed. Additionally, a three-stage transfer learning framework (source-domain pretraining, cross-domain adaptation, and target-domain fine-tuning) improves generalization. Experiments on 8,250 field-collected and augmented images show BMDNet-YOLO achieves 95.6% mAP@0.5, 98.27% precision, and 94.36% recall, surpassing existing baselines. This work offers a robust solution for deploying automated blueberry harvesting systems. Full article

Delivery platforms in urban logistics connect providers with customers through distribution riders, who are usually distinguished by low incomes and limited social rights. This paper aims to compare and analyze the freelance and employment models for riders in different European countries in terms of social sustainability, i.e., work motivation and labor rights. To reach this goal, two activities were performed. On the one hand, qualitative interviews with German and Italian riders were carried out. On the other hand, a dynamic metaheuristic algorithm was developed and implemented to simulate an employment model with a central provider that manages order requests in real-time. The qualitative interviews indicate that riders’ motivations differ between freelance riders and employed riders: freelance riders do feel more controlled. Using a quantitative algorithm, this manuscript shows that when an efficient centralized order–rider assignment strategy is applied, a socially sustainable and simultaneously profitable employment model for food delivery businesses is possible. The results have the potential to legitimize adequate rights and salaries for riders while allowing digital platforms to operate profitably. Such win–win situations could support the implementation of platform structures across different logistics sectors and overcome conflicts regarding working rights in such contexts. Full article

JP-10 (exo-tetrahydrodicyclopentadiene) is a high-energy-density hydrocarbon broadly used in advanced aerospace propulsion as a regenerative cooling fluid; in this study, we aimed to clarify how fuel pressure affects its thermal degradation (oxidative and pyrolytic) in near-isothermal flowing reactor. Experiments were performed under oxidative conditions (wall temperature 623.15 K, p = 0.708–6.816 MPa) and pyrolytic conditions (wall temperature 793.15 K, p = 2.706–7.165 MPa); carbon deposits were quantified by LECO analysis, oxidation activity was assessed by temperature-programmed oxidation (TPO), and morphology was performed by FESEM and EDS. Results show that oxidative coking is minimal (5.37–14.95 μg·cm²) and largely insensitive to pressure in the liquid phase (1.882–6.816 MPa), whereas at 0.708 MPa (gas/phase-change conditions), deposition increases, implicating phase and local heat-transfer effects. Under oxidative conditions, deposits are predominantly amorphous carbon with a disordered structure, formed at relatively low temperatures, with only a few fiber-like metal sulfides identified by EDS. In contrast, under pyrolysis conditions, the deposits are predominantly carbon nanotubes, exhibiting well-defined tubular morphology formed at elevated temperatures via metal-catalyzed growth. The pyrolysis coking yield is substantially higher (66.88–221.89 μg·cm⁻²) and increases with pressure. The findings imply that the pressure influences the coking of JP-10 via phase state under oxidative conditions and residence time under pyrolytic conditions, while basic morphologies of coke deposits remain similar; operationally, maintaining the working pressure higher than the saturated vapor pressure can mitigate oxidation coking associated with phase transitions, and minimizing residence time can mitigate pyrolytic coking. Full article

Multifunctional coatings integrating high transparency, thermal insulation, and self-cleaning properties are critically needed for optical devices and energy-saving applications, yet simultaneously optimizing these functions remains challenging due to material and structural limitations. This study designed a superhydrophobic transparent thermal insulation coating via synergistic co-construction of micro–nano structures using antimony-doped tin oxide (ATO) and SiO₂ nanoparticles dispersed in an epoxy resin matrix, with surface modification by perfluorodecyltriethoxysilane (PFDTES) and γ-glycidyl ether oxypropyltrimethoxysilane (KH560). The optimal superhydrophobic transparent thermal insulating (SHTTI) coating, prepared with 0.6 g SiO₂ and 0.8 g ATO (SHTTI-0.6-0.8), achieved a water contact angle (WCA) of 162.4°, sliding angle (SA) of 3°, and visible light transmittance of 72% at 520 nm. Under simulated solar irradiation, it reduced interior temperature by 7.3 °C compared to blank glass. The SHTTI-0.6-0.8 coating demonstrated robust mechanical durability by maintaining superhydrophobicity through 40 abrasion cycles, 30 tape-peel tests, and sand impacts, combined with chemical stability, effective self-cleaning capability, and exceptional anti-icing performance that prolonged freezing time to 562 s versus 87 s for blank glass. This work provides a viable strategy for high-performance multifunctional coatings through rational component ratio optimization. Full article

This study advances toward establishing the theoretical foundations of Industry 6.0 by developing a comprehensive framework that integrates artificial intelligence (AI), decentralized control systems, and cyber–physical production environments for intelligent, sustainable, and adaptive manufacturing. The research employs a tri-modal methodology (deductive, inductive, and abductive reasoning) to construct a theoretical architecture grounded in five interdependent constructs: advanced technology integration, decentralized organizational structures, mass customization and sustainability strategies, cultural transformation, and innovation enhancement. Unlike prior conceptualizations of Industry 6.0, the proposed framework explicitly emphasizes the cyclical feedback between innovation and organizational design, as well as the role of cultural transformation as a binding element across technological, organizational, and strategic domains. The resulting framework demonstrates that AI-driven decentralized control systems constitute the cornerstone of Industry 6.0, enabling autonomous real-time decision-making, predictive zero-defect manufacturing, and strategic organizational agility through distributed intelligent control architectures. This work contributes foundational theory and actionable guidance for transitioning from centralized control paradigms to AI-driven distributed intelligent manufacturing control systems, establishing a conceptual foundation for the emerging Industry 6.0 paradigm. Full article

Natural latex harvest is pushing the boundaries of unmanned agricultural production in rubber milk collection via integrated robots in hilly and mountainous regions, such as the fixed and mobile tapping robots widely deployed in forests. As there are bad working conditions and complex natural environments surrounding rubber trees, the real-time and precision assessment of rubber milk yield status has emerged as a key requirement for improving the efficiency and autonomous management of these kinds of large-scale automatic tapping robots. However, traditional manual rubber milk yield status detection methods are limited in their ability to operate effectively under conditions involving complex terrain, dense forest backgrounds, irregular surface geometries of rubber milk, and the frequent occlusion of rubber milk bowls (RMBs) by vegetation. To address this issue, this study presents an unmanned aerial vehicle (UAV) imagery rubber milk yield state detection method, termed YOLOv8n-RMB, in unstructured field environments instead of manual watching. The proposed method improved the original YOLOv8n by integrating structural enhancements across the backbone, neck, and head components of the network. First, a receptive field attention convolution (RFACONV) module is embedded within the backbone to improve the model’s ability to extract target-relevant features in visually complex environments. Second, within the neck structure, a bidirectional feature pyramid network (BiFPN) is applied to strengthen the fusion of features across multiple spatial scales. Third, in the head, a content-aware dynamic upsampling module of DySample is adopted to enhance the reconstruction of spatial details and the preservation of object boundaries. Finally, the detection framework is integrated with the BoT-SORT tracking algorithm to achieve continuous multi-object association and dynamic state monitoring based on the filling status of RMBs. Experimental evaluation shows that the proposed YOLOv8n-RMB model achieves an AP@0.5 of 94.9%, an AP@0.5:0.95 of 89.7%, a precision of 91.3%, and a recall of 91.9%. Moreover, the performance improves by 2.7%, 2.9%, 3.9%, and 9.7%, compared with the original YOLOv8n. Plus, the total number of parameters is kept within 3.0 million, and the computational cost is limited to 8.3 GFLOPs. This model meets the requirements of yield assessment tasks by conducting computations in resource-limited environments for both fixed and mobile tapping robots in rubber plantations. Full article

Micron-sized near-spherical α-Al₂O₃ powders are widely used as thermal fillers due to their high thermal conductivity, high packing density, good flowability, and low cost. During the high-temperature calcination, the resulting α-Al₂O₃ powders often exhibit an aggregated worm-like morphology owing to limitations in solid-state mass transfer. Researchers have employed various mineralizers to regulate the morphology of α-Al₂O₃ powders; however, the preparation of micron-sized highly spherical α-Al₂O₃ powders via solid-state calcination is still a great challenge. In this work, micron-sized near-spherical α-Al₂O₃ powders were synthesized through high-temperature calcination using hydratable alumina (ρ-Al₂O₃) as precursor with water-soluble mineralizer ammonium fluoroborate (NH₄BF₄). ρ-Al₂O₃ can undergo a hydration reaction with water to form AlO(OH) and Al(OH)₃ intermediates, serving as an excellent precursor. With the addition of 0.1 wt% NH₄BF₄, the product exhibits an optimal near-spherical morphology. Excessive addition (>0.2wt%), however, significantly promotes the transformation of α-Al₂O₃ from a near-spherical to a plate-like structure. Further studies reveal that the introduction of NH₄BF₄ not only modulates the crystal morphology but also effectively reduces the content of sodium impurities in the powder through a high-temperature volatilization mechanism, thereby enhancing the thermal conductivity of the powder. It is shown that the thermal conductivity of the micron-sized α-Al₂O₃/ epoxy resin composites reaches 1.329 ± 0.009 W/(m·K), which is 7.4 times that of pure epoxy resin. Full article

Background. Physical activity enhances cognitive performance in adolescents, yet the role of circadian timing within the school day remains poorly understood. Purpose. This study examined whether the timing of school-based physical activity (morning, midday, afternoon) influences cognitive performance, subjective alertness, and mood states in early adolescents. Methods. A 12-week crossover intervention was conducted with 102 students (aged 12–13 years) from southern Italy. Each class participated in three 4-week conditions of structured physical activity scheduled in the morning (8:10–9:10), midday (12:10–13:10), and afternoon (15:10–16:10), separated by one-week washouts. Cognitive outcomes (d2-R, Digit Span backward, TMT-A), subjective alertness (KSS), and mood (PANAS-C) were assessed at baseline and after each condition. Analyses employed linear mixed-effects models and repeated-measures ANOVAs, adjusting for sex, BMI, chronotype, and sleep duration. Results. Morning activity produced the strongest improvements in attention (d2-R, η²p = 0.16), working memory (Digit Span backward, η²p = 0.06), processing speed (TMT-A, η²p = 0.08), alertness (KSS, η²p = 0.19), and positive affect (PANAS-C, η²p = 0.05). Midday sessions yielded moderate benefits (d2-R, η²p = 0.09; Digit Span backward, η²p = 0.05; TMT-A, η²p = 0.07; KSS, η²p = 0.09), while afternoon activity showed the weakest or nonsignificant changes (all η²p < 0.05). Chronotype moderated the effects on attention and working memory, with morning types deriving the largest gains. Conclusions. The timing of physical activity is a critical determinant of its cognitive and affective benefits. Incorporating morning exercise into school timetables may represent a low-cost, scalable strategy to optimize both learning readiness and well-being in adolescents. Full article

Erosion wear is a major cause of surface degradation in metallic materials exposed to harsh marine environments. In this study, the erosion wear resistance of the 18Ni300 maraging steel repaired by underwater direct metal deposition (UDMD) is investigated. Results show that UDMD is successfully applied to repair the 18Ni300 samples in underwater environment. Full groove filling and sound metallurgical bonding without cracks are achieved, demonstrating its potential for underwater structural repair. Microstructural analyses reveal good forming quality with fine cellular structures and dense lath martensite in the deposited layer, attributed to rapid solidification under water cooling. Compared to in-air DMD, the UDMD sample exhibits higher surface microhardness due to increased dislocation density and microstructural refinement. Erosion wear behavior is evaluated at 30° and 90° impingement angles, showing that wear mechanisms shift from micro-cutting and plowing at 30° to indentation, crack propagation, and spallation at 90°. The UDMD samples demonstrate superior erosion wear resistance with lower mass loss, particularly at 30°, benefiting from surface work hardening and microstructural advantages. Progressive surface hardening occurs during erosion due to severe plastic deformation, reducing wear rates over time. The combination of refined microstructure, high dislocation density, and enhanced work hardening capability makes UDMD-repaired steel highly resistant to erosive degradation. These findings confirm that UDMD is a promising technique for repairing marine steel structures, offering enhanced durability and long-term performance in harsh offshore environments. Full article

In this study, the behavior of the spherical bearing component of the L-100 bridge part (AlfaTech LLC, Perm, Russia) is considered within the framework of a finite element model. The influence of the pattern of the coupling of the antifriction interlayer with the lower steel plate on the operation of the part is examined in terms of ideal contact, full adhesion, and frictional contact. The thickness of the antifriction interlayer varied from 4 to 12 mm. The dependencies of the contact parameters and the stress–strain state on the thickness were determined. Structurally modified polytetrafluoroethylene (PTFE) without AR-200 fillers was considered the material of the antifriction interlayer. The gradual refinement of the behavioral model of the antifriction material to account for structural and relaxation transitions was carried based on a wide range of experimental studies. The elastic–plastic and primary viscoelastic models of material behavior were constructed based on a series of homogeneous deformed-state experiments. The viscoelastic model of material behavior was refined using data from dynamic mechanical analysis over a wide temperature range [−40; +80] °C. In the first approximation, a model of the deformation theory of plasticity with linear elastic volumetric compressibility was identified. As a second approximation, a viscoelasticity model for the Maxwell body was constructed using Prony series. It was established that the viscoelastic model of the material allows for obtaining data on the behavior of the part with an error of no more than 15%. The numerical analog of the construction in an axisymmetric formulation can be used for the predictive analysis of the behavior of the bearing, including when changing the geometric configuration. Recommendations for the numerical modeling of the behavior of antifriction layer materials and the coupling pattern of the bearing elements are given in this work. A spherical bearing with an antifriction interlayer made of Arflon series material is considered for the first time. Full article

The production of green hydrogen by microalgae is a promising strategy to convert energy of sun light into a carbon-free fuel. Many problems must be solved before large-scale industrial applications. One solution is to find a microalgal species that is easy to grow, easy to manipulate, and that can produce hydrogen open-air, thus in the presence of oxygen, for periods of time as long as possible. In this work we investigate by means of predictive computational models, the [FeFe] hydrogenase enzyme of Nannochloropsis salina, a promising microcalga already used to produce high-value products in salt water. Catalysis of water reduction to hydrogen by [FeFe] hydrogenase occurs in a peculiar iron-sulfur cluster (H-cluster) contained into a conserved H-domain, well represented by the known structure of the single-domain enzyme in Chlamydomonas reinhardtii (457 residues). By combining advanced deep-learning and molecular simulation methods we propose for N. salina a two-domain enzyme architecture hosting five iron-sulfur clusters. The enzyme organization is allowed by the protein size of 708 residues and by its sequence rich in cysteine and histidine residues mostly binding Fe atoms. The structure of an extended F-domain, containing four auxiliary iron-sulfur clusters and interacting with both the reductant ferredoxin and the H-domain, is thus predicted for the first time for microalgal [FeFe] hydrogenase. The structural study is the first step towards further studies of the microalga as a microorganism producing pure hydrogen gas. Full article

As the core component of electrified railway power supply systems, the fatigue performance and reliability of catenary support structures are directly related to the operational safety of high-speed railways. To address the problem of structural fatigue damage caused by increasing train speed and high-frequency operation, this study develops a refined finite element model including a support structure, suspension system and support column, and the dynamic response characteristics and fatigue life evolution law under train operation conditions are systematically analyzed. The results show that under the conditions of 250 km/h speed and 100 times daily traffic, the fatigue lives of the limit locator and positioning support are 43.56 years and 34.48 years, respectively, whereas the transverse cantilever connection and inclined cantilever have infinite life characteristics. When the train speed increases to 400 km/h, the annual fatigue damage of the positioning bearing increases from 0.029 to 0.065, and the service life is shortened by 55.7% to 15.27 years, which proves that high-speed working conditions significantly aggravate the deterioration of fatigue in the structure. The reliability analysis based on Monte Carlo simulation reveals that when the speed is 400 km/h and the daily traffic is 130 times, the structural reliability shows an exponential declining trend with increasing service life. If the daily traffic frequency exceeds 130, the 15-year reliability decreases to 92.5%, the 20-year reliability suddenly decreases to 82.4%, and there is a significant inflection point of failure in the 15–20 years of service. Considering the coupling effect of environmental factors (wind load, temperature and freezing), the actual failure risk may be higher than the theoretical value. On the basis of these findings, engineering suggestions are proposed: for high-speed lines with a daily traffic frequency of more than 130 times, shortening the overhaul cycle of the catenary support structure to 7–10 years and strengthening the periodic inspection and maintenance of positioning support and limit locators are recommended. The research results provide a theoretical basis for the safety assessment and maintenance decision making of high-speed railway catenary systems. Full article

Traditionally, duct sizing in ventilation systems is based on balancing pressure losses across all branches, with fan selection performed subsequently. However, this sequential approach is inadequate for systems with distributed fans in the central duct network, where pressure losses can vary significantly. Consequently, when designing the system topology, fan placement and duct sizing must be considered together. Recent research has demonstrated that discrete optimisation methods can account for multiple load cases and produce ventilation layouts that are both cost- and energy-efficient. However, existing approaches usually concentrate on component placement and assume that duct sizing has already been finalised. While this is sufficient for later design stages, it is unsuitable for the early stages of planning, when numerous system configurations must be evaluated quickly. In this work, we present a novel methodology that simultaneously optimises duct sizing, fan placement, and volume flow controller configuration to minimise life-cycle costs. To achieve this, we exploit the structure of the problem and formulate a mixed-integer linear program (MILP), which, unlike existing non-linear models, significantly reduces computation time while introducing only minor approximation errors. The resulting model enables fast and robust early-stage planning, providing optimal solutions in a matter of seconds to minutes, as demonstrated by a case study. The methodology is demonstrated on a case study, yielding an optimal configuration with distributed fans in the central fan station and achieving a 5 reduction in life-cycle costs compared to conventional central designs. The MILP formulation achieves these results within seconds, with linearisation errors in electrical power consumption below 1.4%, confirming the approach’s accuracy and suitability for early-stage planning. Full article