Search Results (9,835)

With the significant increase in solar power generation’s proportion in power systems, the uncertainty of its power output poses increasingly severe challenges to grid operation. In recent years, solar forecasting models have achieved remarkable progress, with various developed models each exhibiting distinct advantages and characteristics. To address complex and variable geographical and meteorological conditions, it is necessary to adopt a multi-model fusion approach to leverage the strengths and adaptability of individual models. This paper proposes a photovoltaic power prediction framework based on multi-stage ensemble learning, which enhances prediction robustness by integrating the complementary advantages of heterogeneous models. The framework employs a three-level optimization architecture: first, a recursive feature elimination (RFE) algorithm based on LightGBM–XGBoost–MLP weighted scoring is used to screen high-discriminative features; second, mutual information and hierarchical clustering are utilized to construct a heterogeneous model pool, enabling competitive intra-group and complementary inter-group model selection; finally, the traditional static weighting strategy is improved by concatenating multi-model prediction results with real-time meteorological data to establish a time-period-based dynamic weight optimization module. The performance of the proposed framework was validated across multiple dimensions—including feature selection, model screening, dynamic integration, and comprehensive performance—using measured data from a 75 MW photovoltaic power plant in Inner Mongolia and the open-source dataset PVOD. Full article

Remaining useful life (RUL) prediction is a core technology in prognostics and health management (PHM), crucial for ensuring the safe and efficient operation of modern industrial systems. Although deep learning methods have shown potential in RUL prediction, they often face two major challenges: an insufficient generalization ability when distribution gaps exist between training data and real-world application scenarios, and the difficulty of comprehensively capturing complex equipment degradation processes with single-modal data. A key challenge in current research is how to effectively fuse multimodal data and leverage transfer learning to address RUL prediction in small-sample and cross-condition scenarios. This paper proposes an innovative deep multimodal fine-tuning regression (DMFR) framework to address these issues. First, the DMFR framework utilizes a Convolutional Neural Network (CNN) and a Transformer Network to extract distinct modal features, thereby achieving a more comprehensive understanding of data degradation patterns. Second, a fusion layer is employed to seamlessly integrate these multimodal features, extracting fused information to identify latent features, which are subsequently utilized in the predictor. Third, a two-stage training algorithm combining supervised pre-training and fine-tuning is proposed to accomplish transfer alignment from the source domain to the target domain. This paper utilized the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) turbine engine dataset publicly released by NASA to conduct comparative transfer experiments on various RUL prediction methods. The experimental results demonstrate significant performance improvements across all tasks. Full article

Special forces operators are increasingly being utilized as the weapon of choice by many governments on the geopolitical stage. Given the specialized and high-risk nature of special forces operations, it is important to understand the differences that exist when comparing the psychological traits of these groups to regular military forces. An understanding of these traits is crucial when looking to select, develop, and support the most appropriate individuals to succeed in these roles. While previous research has painted a clear picture relating to personality differences between special forces operators and the wider military forces, there is still little research that has explored the psychological traits that both influence and determine performance. As a result, the aim of this study was to explore the perceptions of former United Kingdom (UK) special forces operators regarding the psychological traits they believed were crucial for success as a special forces operator in the UK military. Participants in this study were 20 former UK special forces operators, each having transitioned from active service to civilian life within the previous five years. Data were collected and analyzed using reflexive thematic analysis. Results suggested a specific profile of UK special forces operators composed of nine specific factors: resilience, adaptability, self-belief, perseverance, emotional regulation, self-control, drive, humility, and stubbornness. With the last two relatively novel compared with relevant research in similar populations. These findings can help to underpin the development of special forces-specific programs of support and development. Full article

Background: Tumor-educated platelets (TEPs) represent a promising biosource for non-invasive multi-cancer early detection (MCED). While machine learning (ML) has been applied to TEP data, the integration of explainability to reveal gene-level contributions and regulatory associations remains underutilized. This study aims to develop an interpretable ML framework for cancer detection using platelet RNA-sequencing data, combining predictive performance with biological insight. Methods: This study analyzed 2018 TEP RNA samples from 18 tumor types using seven machine learning classifiers. SHAP (Shapley Additive Explanations) was applied for model interpretability, including global feature ranking, local explanation, and gene-level dependence patterns. A weighted SHAP consensus was built by combining model-specific contributions scaled by Area Under the Receiver Operating Characteristic Curve (AUC). Regulatory insights were supported through network analysis using GeneMANIA. Results: Neural models, including shallow Neural Network (NN) and Deep Neural Network (DNN) achieved the best performance (AUC ~0.93), with Extreme Gradient Boosting (XGB) and Support Vector Machine (SVM) also performing well. Early-stage cancers were predicted with high accuracy. SHAP analysis revealed consistent top features (e.g., SLC38A2, DHCR7, IFITM3), while dependence plots uncovered conditional gene interactions involving USF3 (KIAA2018), ARL2, and DSTN. Multi-hop pathway tracing identified NFYC as a shared transcriptional hub across multiple modulators. Conclusions: The integration of interpretable ML with platelet RNA data revealed robust biomarkers and context-dependent regulatory patterns relevant to early cancer detection. The proposed framework supports the potential of TEPs as a non-invasive, information-rich medium for early cancer screening. Full article

Background/Objectives: To assess the impact of the 2023 FIGO staging revision on stage distribution, survival outcomes, and prognostic performance in endometrial cancer compared to the 2009 system. Methods: This retrospective cohort study analyzed 2969 patients with FIGO 2009 stage I–III endometrial cancer diagnosed at Samsung Medical Center (1994–2023). Patients were reclassified per the 2023 FIGO system. Stage migration, progression-free survival (PFS), and overall survival (OS) were evaluated. Prognostic performance was compared using the Akaike information criterion (AIC), Bayesian information criterion (BIC), concordance index (C-index), and area under the receiver operating characteristic curve (AUC). Results: Stage migration occurred in 20.2% of patients, with 98.3% involving upstaging from FIGO 2009 stage I, largely due to the inclusion of aggressive histology, p53 abnormality, and substantial lymphovascular space invasion (LVSI). The proportion of stage I tumors decreased from 81.5% to 65.2%, while stage II increased to 21.9%, including 14.8% newly classified as stage IIC. Patients remaining in stage I showed favorable outcomes (5-year PFS: 95.3%, OS: 98.5%), whereas those upstaged—especially to stage IIC—had significantly worse outcomes (5-year PFS: 76.5%, OS: 83.1%). Tumors with p53 abnormalities had poorer survival (PFS: 70.8%, OS: 76.6%). The 2023 FIGO system outperformed the 2009 system in prognostic discrimination across all metrics. Conclusions: The FIGO 2023 staging revision improves prognostic accuracy in endometrial cancer by integrating histopathologic and molecular risk factors. These refinements enhance risk stratification and may support more individualized treatment strategies. Full article

Advances in satellite miniaturisation have led to a steep rise in the number of Earth-observation platforms, turning the downlink of the resulting high-volume remote-sensing data into a critical bottleneck. Low-Earth-Orbit (LEO) communication constellations offer a high-throughput relay for these data, yet also introduce intricate scheduling requirements. We term the associated task the Remote Sensing Data Transmission in Communication Constellations (DTIC) problem, which comprises two sequential stages: inter-satellite routing, and satellite-to-ground delivery. This problem can be cast as a Hybrid Flow Shop Scheduling Problem (HFSP). Unlike the classical HFSP, every processor (e.g., ground antenna) in DTIC can simultaneously accommodate multiple jobs (data packets), subject to two-dimensional spatial constraints. This gives rise to a new variant that we call the Hybrid Flow Shop Problem with Two-Dimensional Processor Space (HFSP-2D). After an in-depth investigation of the characteristics of this HFSP-2D, we propose a constructive heuristic, denoted NEHedd-2D, and a Two-Stage Memetic Algorithm (TSMA) that integrates an Inter-Processor Job-Swapping (IPJS) operator and an Intra-Processor Job-Swapping (IAJS) operator. Computational experiments indicate that when TSMA is initialized with the solution produced by NEHedd-2D, the algorithm attains the optimal solutions for small-sized instances and consistently outperforms all benchmark algorithms across problems of every size. Full article

Accurate genome binning is essential for resolving microbial community structure and functional potential from metagenomic data. However, existing approaches—primarily reliant on tetranucleotide frequency (TNF) and abundance profiles—often perform sub-optimally in the face of complex community compositions, low-abundance taxa, and long-read sequencing datasets. To address these limitations, we present MBGCCA, a novel metagenomic binning framework that synergistically integrates graph neural networks (GNNs), contrastive learning, and information-theoretic regularization to enhance binning accuracy, robustness, and biological coherence. MBGCCA operates in two stages: (1) multimodal information integration, where TNF and abundance profiles are fused via a deep neural network trained using a multi-view contrastive loss, and (2) self-supervised graph representation learning, which leverages assembly graph topology to refine contig embeddings. The contrastive learning objective follows the InfoMax principle by maximizing mutual information across augmented views and modalities, encouraging the model to extract globally consistent and high-information representations. By aligning perturbed graph views while preserving topological structure, MBGCCA effectively captures both global genomic characteristics and local contig relationships. Comprehensive evaluations using both synthetic and real-world datasets—including wastewater and soil microbiomes—demonstrate that MBGCCA consistently outperforms state-of-the-art binning methods, particularly in challenging scenarios marked by sparse data and high community complexity. These results highlight the value of entropy-aware, topology-preserving learning for advancing metagenomic genome reconstruction. Full article

Digital twin is a mathematical model that virtually represents a physical object or process and predicts its behavior at future time points. These simulation models enable a deeper understanding of tumorigenic processes and improve biomarker discovery in cancer research. Tumor microenvironment is marked by dysregulated signaling pathways, where kinases and phosphatases serve as critical regulators and promising sources for biomarker discovery. These enzymes operate within multiscale and context-dependent processes where spatial and temporal coordination determine cellular outcomes. Digital Twin technology provides a platform for multimodal and multiscale modeling of kinase and phosphatase processes at the patient-specific level. These models have the potential to transform biomarker validation processes, enhance the prediction of therapeutic responses, and support precision decision-making. In this review, we present the major alterations affecting kinases and phosphatase functions within the tumor microenvironment and their clinical relevance as biomarkers, and we address how digital twins in oncology can augment and refine each stage of the biomarker discovery pipeline. Introducing this emerging technology for cancer biomarker discovery will assist in accelerating its adoption and translation into precision diagnostics and targeted therapies. Full article

This two-part study assesses the impact of biogas inhibition on large-scale waste-to-energy anaerobic digestion (WtE-AD) plants through techno-economic and life cycle assessment approaches. The first part addresses technical and economic aspects. An anaerobic co-digestion system using vegetable waste (FVW) and meat waste (MW) was operated at laboratory scale in a semi-continuous regime with daily feeding to establish a stable process and induce programmed failures causing methanogenic inhibition, achieved by removing MW from the reactor feed and drastically reducing the protein content. Experimental data, combined with bioprocess scale-up models and cost engineering methods, were then used to evaluate the effect of inhibition periods on the profitability of large-scale WtE-AD processes. In the experimental stage, the stable process achieved a yield of 521.5 ± 21 mL CH₄ g⁻¹ volatile solids (VS) and a biogas productivity of 0.965 ± 0.04 L L⁻¹ d⁻¹ (volume of biogas generated per reactor volume per day), with no failure risk detected, as indicated by the volatile fatty acids/total alkalinity ratio (VFA/TA, mg VFA L⁻¹/mg L⁻¹) and the VFA/productivity ratio (mg VFA L⁻¹/L L⁻¹ d⁻¹), both recognized as effective early warning indicators. However, during the inhibition period, productivity decreased by 64.26 ± 11.81% due to VFA accumulation and gradual TA loss. With the progressive reintroduction of the FVW:MW management and the addition of fresh inoculum to the reaction medium, productivity recovered to 96.7 ± 1.70% of its pre-inhibition level. In WtE-AD plants processing 60 t d⁻¹ of waste, inhibition events can reduce net present value (NPV) by up to 40.2% (from 0.98 M USD to 0.55 M USD) if occurring once per year. Increasing plant capacity (200 t d⁻¹), combined with higher revenues from waste management fees (99.5 USD t⁻¹) and favorable electricity markets allowing higher selling prices (up to 0.23 USD kWh⁻¹), can enhance resilience and offset inhibition impacts without significantly compromising profitability. These findings provide policymakers and industry stakeholders with key insights into the economic drivers influencing the competitiveness and sustainability of WtE-AD systems. Full article

The C₄F₇N eco-friendly switchgear shows significant application potential, and quantifying its carbon footprint can accelerate the low-carbon transition in the power industry. A life cycle assessment (LCA) model for a 12 kV C₄F₇N eco-friendly switchgear is established in this study, and the carbon footprint across four stages—raw material acquisition, transportation, operation, and recycling—is accurately quantified. Sensitivity analysis of key raw material parameters and Monte Carlo simulation are used to further quantify the impact of uncertainty in these key sensitive parameters. Results indicate that the operational stage contributes the most to the switchgear’s carbon footprint, amounting to 24,794.77 kgCO₂e, mainly due to electricity consumption. Within this stage, C₄F₇N gas leakage contributes minimally at 2.21 kgCO₂e. The raw material acquisition stage follows with 3005.57 kgCO₂e, where C₄F₇N gas, aluminum, and stainless steel are identified as the primary contributing materials. Sensitivity analysis shows that electricity, C₄F₇N, aluminum, and stainless steel are the resources that have the greatest impact on the switchgear’s carbon footprint. Compared with traditional SF₆ switchgear, the C₄F₇N switchgear has a 23.8% lower total carbon footprint, with its total carbon footprint reaching 26,771.58 kgCO₂e compared to 35,136.48 kgCO₂e for SF₆ switchgear. This advantage stems largely from C₄F₇N’s much lower global warming potential—2090 versus 25,200 for SF₆—which reduces gas-related emissions by 96.6%. These findings substantiate the practical viability of C₄F₇N-based eco-friendly switchgear and provide strategies for the power sector to achieve a low-carbon transition. Full article

Wave energy is a highly concentrated energy resource with five times higher energy density than wind and at least ten times the power density of solar energy. It is expected to make a major contribution to addressing climate change and to help end our dependency on fossil fuels. Many ingenious wave energy conversion methods have been put forward, and a large number of wave energy converters (WECs) have been developed. However, to date, wave energy conversion technology is still in the demonstration application stage. Key issues such as survivability, reliability, and efficient conversion still need to be solved. The major hurdle is the fact that ocean waves provide a slow-moving, high-magnitude force, whereas most electric generators operate at high rotary speed and low torque. Coupling the slow-moving, high-magnitude force of ocean waves normally requires conversion to a high-speed, low-magnitude force as an intermediate step before a rotary generator is applied. This, in general, tends to severely limit the overall efficiency and reliability of the converter and drives the capital cost of the converter well above an acceptable commercial target. Magnetohydrodynamic (MHD) wave energy conversion makes use of MHD generators in which a conducting fluid passes through a very strong magnetic field to produce an electric current. In contrast to alternatives, the relatively slow speed at which the fluid traverses the magnetic field makes it possible to directly couple to ocean waves with a high-magnitude, slowly moving force. The MHD generator provides an excellent match to the mechanical impedance of an ocean wave, and therefore, an MHD WEC has no rotating mechanical parts with high speeds, no complex control process, and has good response to low sea states and high efficiency under all working conditions. This review introduces the system composition, working process, and technical features of WECs based on MHD generators first. Then, the research development, key points, and issues of wave energy conversion technology based on MHD generators are presented in detail. Finally, the problems to be solved and the future research directions of wave energy conversion based on MHD generators are pointed out. Full article

The increasing integration of renewable energy sources and the widespread adoption of electric vehicles have introduced considerable uncertainty into the operation of large-scale power systems. Traditional deterministic unit commitment models are insufficient for managing such variability in a reliable and cost-effective manner. This study proposes a two-stage stochastic unit commitment model that captures uncertainties in solar photovoltaic generation, electric vehicle charging demand, and load fluctuations using a mixed-integer linear programming framework with recourse. The model is applied to Thailand’s national power system, comprising 171 generators across five regions, to assess its scalability for sustainable large-scale planning. Results indicate that the stochastic model significantly enhances system reliability across most demand profiles. Under the Winter Weekday group, the number of lacking scenarios decreases by 76.92 percent and the number of missing periods decreases by 78.57 percent, while the average and maximum lack percentages are reduced by 56.32 percent and 72.61 percent, respectively. Improvements are even greater under the Rainy Weekday group, where lacking scenarios and periods decline by more than 92 percent and the maximum lack percentage falls by over 98 percent, demonstrating the model’s robustness under volatile solar output and load conditions. Although minor anomalies are observed, such as slight increases in average and maximum lack percentages in the Summer Weekday group, these are minimal and likely attributable to randomness in scenario generation or boundary effects in optimization. Overall, the stochastic model provides substantial advantages in managing uncertainty, achieving notable improvements in reliability with only modest increases in operational cost and computational time. The findings confirm that the proposed approach offers a robust and practical framework for supporting sustainable and resilient power systems in regions with high variability in both generation and demand. Full article

Background and Objectives: The extracellular volume fraction (fECV) of the liver, as measured by contrast-enhanced computed tomography (CT), has been shown to correlate closely with the histological stages of hepatic fibrosis. This study aimed to investigate the diagnostic performance of a liver extracellular volume fraction derived from dual-energy CT (DECT) for evaluating liver functional reserve based on the Child–Pugh class in cirrhotic patients, compared with other noninvasive markers. Materials and Methods: This retrospective study included 258 patients with liver cirrhosis who underwent contrast-enhanced DECT. The fECV was measured using iodine maps derived from equilibrium phase images obtained 3 min after contrast injection at 100/140 Sn kVp. Statistical analyses included Welch’s ANOVA with post hoc tests, Spearman’s correlation, and ROC analysis. The area under the curve (AUC) was compared among fECV and other noninvasive markers (aspartate transaminase to platelet ratio index [APRI], Fibrosis-4 [FIB-4], and model for end-stage liver disease [MELD]) using DeLong’s test. Intra- and interobserver reliability of fECV was assessed with the intraclass correlation coefficient (ICC). The area under the receiver operating characteristic curve (AUC) for differentiating Child–Pugh classes was compared among the fECV and other noninvasive markers (aspartate transaminase to platelet ratio index [APRI], Fibrosis-4 [FIB-4], and model for end-stage liver disease [MELD]). Results: The fECV increased significantly with advancing Child–Pugh classes (p < 0.001), showing a moderate correlation with Child–Pugh class (r = 0.53). The mean differences in fECV among the Child–Pugh classes were 8.88 between A and B (95% confidence interval [CI], 5.85–11.92; p < 0.001) and 7.42 between B and C (95% CI, 1.92–12.91: p < 0.001). The AUC for differentiating Child–Pugh classes A and B demonstrated no significant differences among the fECV (0.84), APRI (0.83, p > 0.99) and FIB-4 (0.83, p > 0.99), except for MELD, which had a significantly higher AUC (0.94, p = 0.047). For differentiating Child-Pugh classes B and C, the fECV demonstrated a significantly higher AUC (0.78), compared with FIB-4 (0.50, p = 0.038) and APRI (0.49, p = 0.037), whereas no significant difference was observed between fECV and MELD (0.92, p = 0.12). The intra- and interobserver reliabilities of the fECV measurements were excellent (ICC, 0.93; 95% CI, 0.91–0.95 and 0.91; 95% CI, 0.88–0.92, respectively). Conclusions: DECT derived fECV is a useful noninvasive marker for assessing liver functional reserve based on the Child–Pugh classification. Full article

Battery electric vehicles (BEVs) are gaining market share, yet range anxiety and sparse charging still create demand for hybrids with combustion-engine range extenders. Range-extender vehicles face high customer expectations for noise, vibration, and harshness (NVH) due to their direct comparability with fully electric vehicles. Key challenges include the vibrations of the internal combustion engine, especially from vehicle-induced starts, and the discontinuous operating principle. A technological concept to reduce vibrations in the drivetrain and on the engine mounts, called “FEVcom,” relies on rolling moment compensation. In this concept, a counter-rotating electric machine is coupled to the internal combustion engine via a gear stage to minimize external mount forces. However, due to high speed fluctuations of the crankshaft, the gear drive tends to rattle, which is perceived as disturbing and must be avoided. As part of this work, the rolling moment compensation system was examined regarding its vibration excitation, and an extension to prevent gear rattling was simulated and optimized. For the simulation, the extension, based on a chain or belt drive, was set up as a multi-body simulation model in combination with the range extender and examined dynamically at different speeds. Variations of the extended system were simulated, and recommendations for an optimized layout were derived. This work demonstrates the feasibility of successful rattling avoidance in a range-extender drivetrain with full utilization of the rolling moment compensation. It also provides a solid foundation for further detailed investigations and for developing a prototype for experimental validation based on the understanding gained of the system. Full article

Fatigue in air traffic controllers (ATCOs), particularly within remote tower operations, poses a substantial risk to aviation safety due to its detrimental effects on vigilance, decision-making, and situational awareness. While electroencephalography (EEG) provides a promising avenue for objective fatigue monitoring, existing models often fail to adequately capture both the spatial dependencies across brain regions and the temporal dynamics of cognitive states. To address this challenge, we propose a novel EEG-based fatigue detection framework, Spatio-Temporal Graph with Center Loss Network (STG-CLNet), which jointly models topological brain connectivity and temporal EEG evolution. The model leverages a multi-stage graph convolutional network to encode spatial dependencies and a triple-layer LSTM module to capture temporal progression, while incorporating center loss to enhance feature discriminability in the embedding space. We constructed a domain-specific EEG dataset involving 34 ATCO participants operating in high- and low-traffic remote tower simulations, with fatigue labels derived from three validated subjective metrics. Experimental results demonstrate that STG-CLNet achieves superior classification performance (accuracy = 96.73%, recall = 92.01%, F1-score = 87.15%), outperforming several strong baselines, including LSTM and EEGNet. These findings underscore the potential of STG-CLNet for integration into real-time cognitive monitoring systems in air traffic control, contributing to both theoretical advancement and operational safety enhancement. Full article