Search Results (9,663)

Multi-modal image segmentation is a key task in various fields such as urban planning, infrastructure monitoring, and environmental analysis. However, it remains challenging due to complex scenes, varying object scales, and the integration of heterogeneous data sources (such as RGB, depth maps, and infrared). To address these challenges, we proposed a novel multi-modal segmentation framework, DyFuseNet, which features dynamic adaptive windows and cross-scale feature fusion capabilities. This framework consists of three key components: (1) Dynamic Window Module (DWM), which uses dynamic partitioning and continuous position bias to adaptively adjust window sizes, thereby improving the representation of irregular and fine-grained objects; (2) Scale Context Attention (SCA), a hierarchical mechanism that associates local details with global semantics in a coarse-to-fine manner, enhancing segmentation accuracy in low-texture or occluded regions; and (3) Hierarchical Adaptive Fusion Architecture (HAFA), which aligns and fuses features from multiple modalities through shallow synchronization and deep channel attention, effectively balancing complementarity and redundancy. Evaluated on benchmark datasets (such as ISPRS Vaihingen and Potsdam), DyFuseNet achieved state-of-the-art performance, with mean Intersection over Union (mIoU) scores of 80.20% and 80.65%, surpassing MFTransNet by 1.71% and 1.57%, respectively. The model also demonstrated strong robustness in challenging scenes (such as building edges and shadowed objects), achieving an average F1 score of 85% while maintaining high efficiency (26.19 GFLOPs, 30.09 FPS), making it suitable for real-time deployment. This work presents a practical, versatile, and computationally efficient solution for multi-modal image analysis, with potential applications beyond remote sensing, including smart monitoring, industrial inspection, and multi-source data fusion tasks. Full article

The growing penetration of renewable energy sources requires resilient microgrids capable of providing stable and continuous operation. Hybrid energy storage systems (HESS), which integrate hydrogen-based storage systems (HBSS), battery storage systems (BSS), and supercapacitor banks (SCB), are essential to ensuring the flexibility and robustness of these microgrids. Accurate modelling of these microgrids is crucial for analysis, controller design, and performance optimization, but the complexity of HESS poses a significant challenge: simplified linear models fail to capture the inherent nonlinear dynamics, while nonlinear approaches often require excessive computational effort for real-time control applications. To address this challenge, this study presents a novel state space model with linear variable parameters (LPV), which effectively balances accuracy in capturing the nonlinear dynamics of the microgrid and computational efficiency. The research focuses on a high-voltage DC bus microgrid architecture, in which the BSS and SCB are connected directly in parallel to provide passive DC bus stabilization, a configuration that improves system resilience but has received limited attention in the existing literature. The proposed LPV framework employs recursive linearisation around variable operating points, generating a time-varying linear representation that accurately captures the nonlinear behaviour of the system. By relying exclusively on directly measurable state variables, the model eliminates the need for observers, facilitating its practical implementation. The developed model has been compared with a reference model validated in the literature, and the results have been excellent, with average errors, MAE, RAE and RMSE values remaining below 1.2% for all critical variables, including state-of-charge, DC bus voltage, and hydrogen level. At the same time, the model maintains remarkable computational efficiency, completing a 24-h simulation in just 1.49 s, more than twice as fast as its benchmark counterpart. This optimal combination of precision and efficiency makes the developed LPV model particularly suitable for advanced model-based control strategies, including real-time energy management systems (EMS) that use model predictive control (MPC). The developed model represents a significant advance in microgrid modelling, as it provides a general control-oriented approach that enables the design and operation of more resilient, efficient, and scalable renewable energy microgrids. Full article

Transportation systems are particularly vulnerable to disruptions such as pandemics, which create significant challenges for maintaining efficiency, safety, and service quality. This study focuses on rail passenger transport in the Slovak Republic and develops a simulation framework to evaluate system performance under crisis conditions. Weekly data from the national rail operator for the period 2019–2021 were combined with information on governmental restrictions, standardized into a five-level framework. A nonlinear autoregressive model with exogenous inputs (NARX), implemented and validated in MATLAB R2021b (MathWorks, Natick, MA, USA), was applied to simulate the impact of restrictive measures on passenger demand. The results revealed a strong relationship between the severity of measures and ridership levels, with the most significant effects observed in education, workplace access, movement limitations, and retail. For instance, during complete school closures, passenger volumes declined by up to 75% relative to the pre-pandemic baseline. Based on the simulation outcomes, recommendations were formulated for adapting railway operations, including dynamic adjustments of transport capacity (10–40%) according to restriction levels. The proposed modelling and simulation approach offers transport authorities a cost-effective tool for scenario testing, disruption management, and the design of resilient passenger rail systems capable of adapting to crises and uncertainties. Full article

The ongoing energy transition and decarbonization efforts have prompted the development of Hybrid Renewable Energy Systems (HRES) capable of integrating multiple generation and storage technologies to enhance energy autonomy. Among the available options, hydrogen has emerged as a versatile energy carrier, yet most studies have focused either on stationary applications or on mobility, seldom addressing their integration withing a single framework. In particular, the potential of Metal Hydride (MH) tanks remains largely underexplored in the context of sector coupling, where the same storage unit can simultaneously sustain household demand and provide in-house refueling for light-duty fuel-cell vehicles. This study presents the design and analysis of a residential-scale HRES that combines photovoltaic generation, a PEM electrolyzer, a lithium-ion battery and MH storage intended for direct integration with a fuel-cell electric microcar. A fully dynamic numerical model was developed to evaluate system interactions and quantify the conditions under which low-pressure MH tanks can be effectively integrated into HRES, with particular attention to thermal management and seasonal variability. Two simulation campaigns were carried out to provide both component-level and system-level insights. The first focused on thermal management during hydrogen absorption in the MH tank, comparing passive and active cooling strategies. Forced convection reduced absorption time by 44% compared to natural convection, while avoiding the additional energy demand associated with thermostatic baths. The second campaign assessed seasonal operation: even under winter irradiance conditions, the system ensured continuous household supply and enabled full recharge of two MH tanks every six days, in line with the hydrogen requirements of the light vehicle daily commuting profile. Battery support further reduced grid reliance, achieving a Grid Dependency Factor as low as 28.8% and enhancing system autonomy during cold periods. Full article

In this study, the impact of the electric motor size and the hybridization ratio of a Fuel Cell Electric Bus on its vehicle performance (i.e., gradeability and acceleration) and fuel consumption was investigated using the ADVISOR software. The investigation first involved a parametric analysis with different electric motor and fuel cell sizes for the dynamic performance metrics, specifically the 0–60 km/h vehicle acceleration and the maximum gradeability (%) at a constant speed of 20 km/h. The results revealed that the acceleration is most sensitive to fuel cell power. Regarding gradeability, a more complex relationship was observed: when the electric motor power was below 215 kW, gradeability remained consistently low regardless of the fuel cell size. However, for motors exceeding 215 kW, fuel cell power then became a significant influencing factor on the vehicle’s climbing capability. Subsequently, the analysis focused on the effect of the hybridization ratio, which represents the power balance between the fuel cell and the energy storage system, varied between 0 and 0.8. Results showed that increasing the hybridization ratio decreases gradeability and acceleration performance and increases total energy consumption. This trade-off is quantitatively illustrated by the results over the Central Business District (CBD) driving cycle. For instance, the pure battery-electric configuration (a hybridization ratio of 0), featuring a 296 kW battery system, recorded a gradeability of 12.4% and an acceleration time of 16.3 s, while consuming 28,916 kJ. At an intermediate hybridization ratio of 0.4 (composed of a 118.4 kW fuel cell and a 177.6 kW battery), performance remained high with a gradeability of 12.2% and an acceleration of 17.3 s, but the energy consumption increased to 43,128 kJ. Finally, in the fuel-cell-dominant configuration with a hybridization ratio of approximately 0.8 (a 236.8 kW fuel cell and a 59.2 kW battery), gradeability dropped to 8.4%, acceleration time deteriorated to 38.9 s, and total energy consumption increased further to 52,678 kJ over the CBD driving cycle. Full article

Organoid technology has emerged as a revolutionary tool in cancer research, offering physiologically accurate, three-dimensional models that preserve the histoarchitecture, genetic stability, and phenotypic complexity of primary tumors. These self-organizing structures, derived from adult stem cells, induced pluripotent stem cells, or patient tumor biopsies, recapitulate critical aspects of tumor heterogeneity, clonal evolution, and microenvironmental interactions. Organoids serve as powerful systems for modeling tumor progression, assessing drug sensitivity and resistance, and guiding precision oncology strategies. Recent innovations have extended organoid capabilities beyond static culture systems. Integration with microfluidic organoid-on-chip platforms, high-throughput CRISPR-based functional genomics, and AI-driven phenotypic analytics has enhanced mechanistic insight and translational relevance. Co-culture systems incorporating immune, stromal, and endothelial components now permit dynamic modeling of tumor–host interactions, immunotherapeutic responses, and metastatic behavior. Comparative analyses with conventional platforms, 2D monolayers, spheroids, and patient-derived xenografts emphasize the superior fidelity and clinical potential of organoids. Despite these advances, several challenges remain, such as protocol variability, incomplete recapitulation of systemic physiology, and limitations in scalability, standardization, and regulatory alignment. Addressing these gaps with unified workflows, synthetic matrices, vascularized and innervated co-cultures, and GMP-compliant manufacturing will be crucial for clinical integration. Proactive engagement with regulatory frameworks and ethical guidelines will be pivotal to ensuring safe, responsible, and equitable clinical translation. With the convergence of bioengineering, multi-omics, and computational modeling, organoids are poised to become indispensable tools in next-generation oncology, driving mechanistic discovery, predictive diagnostics, and personalized therapy optimization. Full article

Accurate demand forecasting contributes to improved energy efficiency and the development of short-term strategies. Predictive management of energy storage using redox flow batteries is presented as a robust solution for optimizing the operation of microgrids from the demand side. This study proposes an intelligent architecture that integrates demand forecasting models based on artificial neural networks and active management strategies based on the instantaneous production of renewable sources within the microgrid. The solution is supported by a real-time monitoring platform capable of analyzing data streams using continuous evaluation algorithms, enabling dynamic operational adjustments and active methods for predicting the storage system’s state of charge. The model’s effectiveness is validated using performance indicators such as RMSE, MAPE, and MSE, applied to experimental data obtained in a specialized microgrid laboratory. The results also demonstrate substantial improvements in energy planning and system operational efficiency, positioning this proposal as a viable strategy for distributed and sustainable environments in modern electricity systems. Full article

Precise forecasting of carbon content in the converter’s reblowing phase is pivotal to boosting steel production efficiency and ensuring effective control over molten steel quality. However, existing mechanistic models based on material balance and decarbonization kinetics suffer from insufficient accuracy due to simplifying assumptions. In contrast, data-driven models rely on data quality, lack generalization capability, and lack physical interpretability. Additionally, integral models based on flue gas analysis suffer from data latency issues. To overcome these limitations, this study proposed a real-time carbon content prediction model for the converter’s reblowing phase, leveraging a physics-informed long short-term memory (PI-LSTM) network. First, flue gas data was processed using a carbon integration model to generate a carbon content change curve during the reblowing stage as a reference for actual values; second, a dual-branch network structure was designed, where the LSTM branch simultaneously predicts carbon content and key unmeasurable parameters in the decarbonization kinetics, while the mechanism branch combined these parameters with the decarbonization formula to calculate carbon content under mechanism constraints; finally, a joint loss function (combining data-driven loss and mechanism constraint loss) was used to train the model, and the gray wolf optimization (GWO) algorithm was employed to optimize hyperparameters. Experimental results show that compared to the mechanism model (MM) and LSTM model, the PI-LSTM model achieves an average absolute error (MAE) of 0.0077, a root mean square error (RMSE) of 0.0112, and endpoint carbon content hit rates within ±0.005%, ±0.01%, ±0.015% error ranges, achieving 53.71%, 82.23%, and 95.45%, respectively, significantly improving prediction accuracy and physical plausibility. This model lays a robust groundwork for dynamic closed-loop real-time control of carbon levels in the converter’s reblowing stage. Full article

Object detection has a vital impact on the analysis and interpretation of visual scenes. It is widely utilized in various fields, including healthcare, autonomous driving, and vehicle surveillance. However, complex scenes containing small, occluded, and multiscale objects present significant difficulties for object detection. This paper introduces a lightweight object detection algorithm, utilizing YOLOv8n as the baseline model, to address these problems. Our method focuses on four steps. Firstly, we add a layer for small object detection to enhance the feature expression capability of small objects. Secondly, to handle complex forms and appearances, we employ the C2f-DCNv2 module. This module integrates advanced DCNv2 (Deformable Convolutional Networks v2) by substituting the final C2f module in the backbone. Thirdly, we designed the CBAM, a lightweight attention module. We integrate it into the neck section to address missed detections. Finally, we use Ghost Convolution (GhostConv) as a light convolutional layer. This alternates with ordinary convolution in the neck. It ensures good detection performance while decreasing the number of parameters. Experimental performance on the PASCAL VOC dataset demonstrates that our approach lowers the number of model parameters by approximately 9.37%. The mAP@0.5:0.95 increased by 0.9%, recall (R) increased by 0.8%, mAP@0.5 increased by 0.3%, and precision (P) increased by 0.1% compared to the baseline model. To better evaluate the model’s generalization performance in real-world driving scenarios, we conducted additional experiments using the KITTI dataset. Compared to the baseline model, our approach yielded a 0.8% improvement in mAP@0.5 and 1.3% in mAP@0.5:0.95. This result indicates strong performance in more dynamic and challenging conditions. Full article

This research details the development and evaluation of a Modern Ultrasonic Cleaning Tank (MUCT) designed to enhance cleaning efficiency in jewelry manufacturing, particularly for silver jewelry, replacing the traditional method, which was less efficient and had higher operating costs. The MUCT offers capabilities of single- or dual-frequency ultrasonic operation (28 kHz and 40 kHz) and adjustable transducer positioning. An advanced method involving computer simulations, utilizing harmonic response analysis and transient dynamic analysis, was employed to determine the acoustic pressure inside the MUCT, thereby indicating the cavitation intensity required to achieve high cleaning efficiency. Simulation results confirm that this design can distribute acoustic pressure throughout the MUCT, as intended. A prototype MUCT was assembled, and its operation was validated through foil corrosion tests, ultrasonic power concentration (UPC) measurements, and jewelry cleaning tests. The results revealed that the MUCT’s center provided the maximum UPC of 28 W/L and an acoustic pressure of 30.43 MPa, effectively operating at single and dual frequencies, and achieving superior dirt removal. The highest cleaning efficiency of 100% was achieved using dual frequency with a 97% water and 3% dishwashing liquid mixture at 60 °C, exceeding the 23.52% obtained with water at 27 °C without ultrasonic treatment. The MUCT, successfully integrated into the manufacturing process, offers customizable features to meet various cleaning needs, providing flexibility, improved performance, and cost savings. Full article

Sindbis virus (SINV) is a mosquito-borne alphavirus capable of causing neurological and immunological symptoms in humans, yet its effects on neural/immune systems remain insufficiently characterized. This study aimed to examine SINV replication, UV-C light inactivation, apoptosis induction, and immune gene modulation in human SH-SY5Y neuroblastoma cells. Following viral adaptation and infectious dose determination, SINV replication and inactivation were assessed using RT-qPCR and dsRNA immunofluorescence. Apoptotic markers (caspase-3, Bax, Bcl-2) were analyzed by immunofluorescence and immune genes expression kinetics (TLR3/7, RIGI, MDA5, IL-1β, IL-6, TNFα, IL-10, IFNβ and β-catenin) were measured at defined time points post-infection by RT-qPCR. SH-SY5Y cells supported productive SINV infection, with viral RNA detectable as early as 3 hpi and marked cytopathic effects by 24 hpi. A custom-built UV-C chamber achieved complete viral inactivation following 3 × 30 s exposures. We observed SINV time-course replication and UV-C inactivation with conspicuous morphological alterations in SH-SY5Y cells. Furthermore, SINV triggered caspase-dependent apoptosis and robust transcriptional upregulation of innate immune genes, peaking between 12–16 hpi and declining by 30 hpi. These findings elucidate the temporal dynamics of SINV replication, cell death mechanisms, and immune activation in a neuronal context, contributing to a better understanding of SINV neuropathogenesis. Full article

A hybrid control architecture is proposed for enhancing the stability and energy management of DC microgrids (DCMGs) integrating photovoltaic generation, batteries, and supercapacitors. The approach combines nonlinear Sliding Mode Control (SMC) for fast and robust DC bus voltage regulation with a Deep Q-Learning (DQL) agent that learns optimal high-level policies for charging, discharging, and load management. This dual-layer design leverages the real-time precision of SMC and the adaptive decision-making capability of DQL to achieve dynamic power sharing and balanced state-of-charge levels across storage units, thereby reducing asymmetric wear. Simulation results under variable operating scenarios showed that the proposed method significantly improvedvoltage stability, loweredthe occurrence of deep battery discharges, and decreased load shedding compared to conventional fuzzy-logic-based energymanagement, highlighting its effectiveness and resilience in the presence of renewable generation variability and fluctuating load demands. Full article

This paper proposes an innovative hybrid optimization framework for the optimal installation and operation of thyristor-switched capacitors (TSCs) within medium-voltage distribution networks, targeting both energy losses reduction and cost efficiency. The core of the approach combines the exploratory capabilities of the atan-sinc optimization algorithm (ASOA), a recent metaheuristic inspired by mathematical functions, with the local refinement power of the IPOPT solver within a master–slave architecture. This integrated method addresses the inherent complexity of a multi-objective, mixed-integer nonlinear programming problem that seeks to balance conflicting goals: minimizing annual system losses and investment costs. Extensive testing on IEEE 33- and 69-bus systems under fixed and dynamic reactive power injection scenarios demonstrates that our framework consistently delivers superior solutions when compared to traditional and state-of-the-art algorithms. Notably, the variable operation case yields energy savings of up to 12%, translating into annual monetary gains exceeding USD 1000 in comparison with the fixed support scenario.The solutions produce well-distributed Pareto fronts that illustrate valuable trade-offs, allowing system planners to make informed decisions. The findings confirm that the proposed strategy constitutes a scalable, and robust tool for reactive power planning, supporting the deployment of smarter and more resilient distribution systems. Full article

Stony corals are long-lived, calcifying cnidarians that can be preserved within archaeological strata, offering insights into past seawater conditions, anthropogenic influences, and harbor dynamics. This study analyzes sub-fossil Cladocora sp. colonies from ancient Akko, Israel, dated to the Hellenistic period (~335–94 BCE), alongside modern Cladocora caespitosa from Haifa Bay, Israel. We employed micromorphology, stable isotope analysis, and DNA sequencing to assess species identity, colony growth form, and environmental conditions experienced by the corals. Comparisons suggest that Hellenistic Akko corals grew in high-light, cooler-water, high-energy environments, potentially with exposure to terrestrial waste. The exceptional preservation of these colonies indicates rapid burial, possibly linked to ancient harbor activities or extreme sedimentation. Our results demonstrate the utility of scleractinian corals as valuable paleoenvironmental archives, capable of integrating both biological and geochemical proxies to reconstruct past marine conditions. By linking archaeological and ecological records, this multidisciplinary approach provides a comprehensive understanding of historical coastal dynamics, including ancient harbor use, climate variability, and anthropogenic impacts. Full article

Visible near-infrared (VNIR) spectroscopy offers a cost-effective solution to quantify the spatiotemporal dynamics of soil organic carbon (SOC), especially in the context of rapid advances in spectra-based local modeling approaches using large-scale soil spectral libraries. And yet, direct temporal transferability of VNIR spectroscopic modeling (applying historical models to new spectral data) and its capability to monitor temporal changes in SOC remain underexplored. To address this gap, this study uses the LUCAS Soil dataset (2009 and 2015) from France to evaluate the effectiveness of localized spectral models in detecting SOC changes. Two local learning algorithms, memory-based learning (MBL) and GLOBAL-LOCAL algorithms, were adapted to integrate spectral and soil property similarities during local training set selection, while also incorporating LUCAS 2009 soil measurements (clay, silt, sand, CEC) as covariates. These adapted local learning algorithms were then compared against global partial least squares regression (PLSR). The results demonstrated that localized models substantially outperformed global PLSR, with MBL achieving the highest accuracy for croplands, grasslands, and woodlands (R² = 0.72–0.79, RMSE = 4.73–20.92 g/kg). Incorporating soil properties during the local learning procedure reduced spectral heterogeneity, leading to improved SOC prediction accuracy. This improvement was particularly pronounced after excluding organic soils from grasslands and woodlands, as evidenced by 13.3–21.1% decreases in the RMSE. Critically, for SOC monitoring, spectrally predicted SOC successfully identified over 70% of samples experiencing significant SOC changes (>10% loss or gain), effectively capturing the spatial patterns of SOC changes. This study demonstrated the potential of localized spectral modeling as a cost-effective tool for monitoring SOC dynamics, enabling efficient and large-scale assessments critical for sustainable soil management. Full article