Search Results (4,023)

Reliable and rapid forecasting of wind turbine output power is vital for operators, particularly day-ahead and intraday market scheduling and reserve allocation. However, the inherent unpredictability, intermittency, and volatility of wind turbine output make forecasting processes difficult. To address this challenge, this study proposes a Sinusoidal Representation Network (SIREN)-based forecasting model for high-accuracy, rapid direct multi-horizon forecasting of wind turbine output power. SIREN is selected due to the periodic and symmetrical mathematical structure of its sinusoidal activation function, which allows the model to represent both low-frequency trends and high-frequency sudden changes in wind energy data. To improve data quality, compensate for asymmetric fluctuations in wind data, and provide more suitable inputs for SIREN training. Several preprocessing steps are utilized before feeding the data into the model. The proposed preprocessing step includes a moving median filter, robust scaling based on median and interquartile range, Winsorizing clipping, and a Hampel filter to reduce the effects of instantaneous noise, outliers, and local peaks without disrupting temporal continuity. Subsequently, a Savitzky–Golay smoothing is applied to attenuate high-frequency measurement noise while preserving curvature, local peaks, and physically meaningful short-term dynamics in the data. The sliding-window approach is used to formulate the multi-horizon forecasting problem directly, and a direct h-step-ahead forecasting architecture is designed, preserving structural symmetry in the time series. The SIREN is trained and tested using MATLAB with the help of two different datasets: Dataset-1 has a 10 min resolution for 1 year, and Dataset-2 has a 1 h resolution for 15 years. The forecast horizon parameter h is considered separately for each step, and the proposed SIREN is independently trained, validated, and tested for each target horizon while maintaining chronological order. The results demonstrate that the proposed model is able to yield high forecast performance for a wide spectrum of horizons ranging from 10 min to 15 days. The accuracy of the proposed model for Dataset-1 is R² of 99.6%, MSE of 0.085%, MAE of 1.7%, and MAPE of 12%, while for Dataset-2, the accuracy is R² of 98.8%, MSE of 0.3%, MAE of 3.6%, and MAPE of 23%. Ablation and sensitivity analyses are conducted to evaluate the impact of the basic components used in the proposed model on forecasting performance. In addition, combative experiments are performed using traditional time series, ML, and DL forecasting techniques to better assess the contribution of the model. The obtained results show that the SIREN-based direct forecasting approach provides strong learning capability, as well as high forecasting accuracy, for both high-resolution and low-resolution wind power data. Overall, its ability to capture the symmetric and periodic characteristics inherent in wind turbine power data makes it a promising alternative for multi-horizon wind power forecasting applications. Full article

Virtually all spaceborne topographic lidars to date have used a single beam, with the exception of the ATLAS lidar on NASA’s ICESat-2 satellite, which split the beam into 3 “strong” and 3 “weak” beamlets distributed perpendicular to the along-track path of the satellite. This approach has provided high-resolution along-track surface measurements but relatively poor resolution cross-track measurementswithin a given surface area. The present paper attempts to resolve this discrepancy by (1) transmitting and scanning a single Gaussian beam and (2) imaging the return onto a 14 × 14 pixelated, single-photon sensitive, detector array, thereby providing between 100 and 196 measurements per pulse, depending on the solar background. Besides enhancing the lidar’s capability to penetrate tree canopies and water bodies, the proposed single-beam approach provides one to two orders of magnitude more measurements per pulse with equal spatial resolution in boththe along-track and cross-track directions. At the 10 kHz pulse rate of the ATLAS laser on NASA’s ICESat-2 satellite, this implies between 1 and 2 million topographic measurements per second. The maximum surface area observable by a single pulse increases with the laser peak power defined by the ratio of the pulse energy to the temporal pulsewidth. Larger surface areas per pulse result in more time for cross-track scanning while still maintaining contiguous along-track mapping. Two scanning methods appear to be feasible: (1) circular scans using individual but temporally coordinated wedge scanners for the transmitted and received beams, and (2) unidirectional linear scans utilizing Acousto-Optic Deflectors. The circular scan approach is probably easier to implement, but it also requires additional post-processing to obtain an accurate contiguous 3D image of the planetary terrain. Full article

Does mandatory sustainability disclosure improve the quality of corporate financial reporting immediately, gradually, or with delay? We address this question using Saudi Arabia’s January 2021 Tadawul ESG Disclosure Guidelines—the first comprehensive sustainability disclosure framework in the Gulf Cooperation Council and a uniform, accurately dated regulatory shock affecting all listed firms. Using a balanced panel of 135 non-financial firms over 2017–2024 (1080 firm-year observations), we estimate absolute discretionary accruals from the Modified Jones Model and employ event-time fixed-effects regressions with Driscoll–Kraay standard errors robust to heteroskedasticity, autocorrelation, and cross-sectional dependence. We document a temporal paradox: reporting quality did not change in the announcement year (2021), deteriorated significantly in 2022 (+28%) and 2023 (+38%) relative to the pre-reform baseline, and then improved significantly in 2024 (−17%). The pattern survives performance-matched discretionary accruals, exclusion of the 2020 COVID-19 year, a placebo test, sectoral disaggregation across nine Tadawul-aligned industry groups, and a battery of pre-reform firm characteristics. Heterogeneity analysis identifies the underlying mechanism: voluntary pre-2021 ESG disclosers and firms with stronger pre-reform governance exhibit amplified short-run deterioration, while larger firms with pre-existing reporting infrastructure show a substantially attenuated paradox. These patterns are jointly consistent with the adjustment-cost mechanism we develop: the reform redirected scarce reporting governance toward the new disclosure margin during a three-year compliance buildout, after which the constraining effect on accrual-based earnings management emerged. The findings carry direct implications for the design and evaluation of mandatory sustainability disclosure reforms currently advancing across emerging and developed markets. Full article

Predicting near-future monetization in virtual livestreaming remains methodologically challenging because paid-support events are sparse, temporally dependent, and vulnerable to leakage under inappropriate evaluation designs. This study develops a leakage-guarded, window-based machine-learning framework for predicting next-window Superchat occurrence from VTuber live-chat dynamics. Public VTuber live-chat and Superchat logs were reconstructed into non-overlapping five-minute windows, and features were organized into audience activity, member composition, message intensity, donation-state information, and short-horizon dynamic groups. To reduce optimistic bias, the primary evaluation used video-level grouped splitting and compared a strict setting that excluded direct current-window donation-state variables with an extended donation-state-aware setting. HistGradientBoosting achieved the strongest performance. In the strict setting, it reached PR-AUC = 0.899, ROC-AUC = 0.920, F1 = 0.822, and Brier score = 0.171, while the extended setting produced only modest additional gains. Additional zero-chat sensitivity, repeated grouped split, channel-level robustness, graph-proxy baseline, feature-ablation, and calibration analyses supported the stability and interpretability of the framework. The results suggest that next-window Superchat occurrence can be predicted from participation breadth, chat activity, message intensity, and temporally shifted behavioral dynamics under leakage-aware evaluation. Full article

Background: Acute compartment syndrome is a surgical emergency that is most often associated with trauma. Rarely, however, it can develop without major trauma in patients receiving anticoagulation therapy. Methods: A 53-year-old woman receiving warfarin therapy presented with progressive swelling and pain in the left forearm after physical therapy. Computed tomography angiography showed preserved arterial flow. On clinical examination, pain and paralysis were present, whereas pallor and paresthesia were absent. Direct compartment pressure measurements demonstrated markedly elevated pressures in the dorsal and deep volar compartments, measuring 88 mmHg and 110 mmHg, respectively. Emergency fasciotomy was performed approximately 9 h after symptom onset. Results: Intraoperative findings showed hematoma formation and partial muscle necrosis in the deep volar compartment. After surgery, persistent bleeding required repeated hemostatic interventions and blood-product transfusion. Negative-pressure wound therapy was applied, and delayed primary closure was subsequently performed. Anticoagulation therapy was temporarily discontinued and later resumed. The patient recovered without further bleeding, wound complications, or functional impairment and was discharged in stable condition. No complications, including hematoma recurrence or infection, were observed during 6 months of follow-up. Conclusions: This case highlights a temporal association between physical therapy and acute forearm compartment syndrome in a patient receiving anticoagulation therapy. It also emphasizes that preserved peripheral pulses and intact arterial flow do not exclude the diagnosis. Early recognition and prompt surgical intervention remain essential to prevent irreversible tissue damage and optimize outcomes. Full article

Bilateral deficit (BLD) is traditionally defined as the reduced capacity to produce force during simultaneous bilateral contractions compared with the summed output of unilateral actions. However, in applied sport settings, BLD is frequently estimated from countermovement jump (CMJ) performance, representing a functional rather than a direct mechanical measure of force production. Therefore, the aim of this study was to examine the association between a CMJ-derived functional BLD index and multidirectional performance in soccer players. Forty male university soccer players (age: 23 ± 1 years) performed unilateral and bilateral CMJ. The BLD index was calculated from jump height values obtained during these assessments. Participants subsequently completed the 505 change-of-direction (CoD) test, which was analyzed using two-dimensional video-based motion analysis. Participants were classified according to BLD magnitude into low, moderate, and high BLD groups. Group differences were assessed using Kruskal–Wallis tests with Bonferroni-adjusted post hoc comparisons. Additionally, Spearman correlation analyses were performed using BLD as a continuous variable. Significant between-group differences were observed across all temporal phases of the 505 test (p < 0.001), with players exhibiting lower BLD values demonstrating superior acceleration, deceleration, reacceleration, and overall CoD performance. Significant negative correlations were also observed between BLD and reaction time, acceleration, deceleration, reacceleration, CoD time, and CoD deficit (r_s = −0.42 to −0.69; p < 0.001). No significant associations were found for stride length, acceleration ability, or inter-limb asymmetry. These findings suggest that lower magnitudes of a CMJ-derived functional BLD index are associated with superior multidirectional performance in soccer players. However, given that BLD was estimated from jump performance, the results should be interpreted as associations with a functional neuromuscular performance index rather than as direct evidence of bilateral force production capacity. Full article

The increasing penetration of photovoltaic (PV) systems requires accurate and stable short-term forecasting to ensure reliable grid operation under operational constraints. This paper investigates short-horizon multi-step PV power forecasting using one full year of high-resolution (5 min) real-world data from a 111-kW grid-connected rooftop installation. The forecasting problem is formulated as a direct multi-output supervised learning task with a 30 min prediction horizon. A comprehensive comparative evaluation is conducted across baseline (persistence), tree-based (XGBoost), and deep learning architectures (LSTM, GRU, and Temporal Convolutional Networks—TCN). Results show that deep learning models significantly outperform conventional baselines, with LSTM achieving the lowest normalized RMSE (≈10.3%), while TCN provides a competitive trade-off between predictive accuracy, temporal stability, and computational efficiency. The direct multi-step formulation was adopted to reduce potential error propagation effects commonly observed in recursive forecasting approaches. Beyond forecasting accuracy, the study evaluates computational complexity and inference latency to assess practical deployability in resource-constrained environments. The proposed framework demonstrates that high-resolution real-world PV forecasting can achieve both strong predictive performance and operational feasibility. These findings contribute to the development of robust short-term forecasting strategies for distributed renewable energy systems operating under regulatory export constraints. Full article

Continuous monitoring of seawater alkalinity is essential for maintaining chemical stability in coastal environments and supporting efficient operation of desalination and water-treatment systems; however, direct alkalinity measurements are often limited in temporal resolution. This study develops and evaluates a machine learning framework for estimating seawater alkalinity using quality-controlled daily and sub-daily water-quality observations collected from a coastal monitoring station along the Arabian Gulf coast of eastern Saudi Arabia during 2017–2023. Five machine learning models, Random Forest (RF), Gradient Boosting (GB), Extreme Gradient Boosting (XGB), Support Vector Regression (SVR), and K-Nearest Neighbors (KNN), are assessed under two configurations: a baseline setup relying on the original predictor variables and an enhanced setup incorporating wavelet-decomposed features to represent multiscale temporal variability. Model performance is evaluated using five-fold cross-validation and quantified using R², root mean square error (RMSE), and mean absolute error (MAE). Under the baseline configuration, ensemble-based models outperform single-estimator and distance-based approaches, with RF achieving the best performance (R² = 0.77, RMSE = 2.57 ppm, MAE = 1.71 ppm). The incorporation of wavelet-based feature enrichment leads to consistent performance improvements across all models, reflected by higher R² values and reduced RMSE and MAE. The wavelet-enhanced RF model exhibits the strongest overall performance, attaining a mean R² of approximately 0.91 together with an RMSE of about 1.6 ppm and an MAE of around 1.0 ppm, while also showing reduced variability across cross-validation folds. The XGB model shows notable improvement with wavelet enrichment, whereas SVR and KNN benefit mainly through moderate error reduction. Overall, the findings show that wavelet-based feature enrichment improves the accuracy and stability of ML models for seawater alkalinity estimation, with RF providing the most reliable performance for coastal monitoring applications. Full article

Floating wind turbines (FWTs) are key equipment for deep-sea clean energy exploitation, and their structural health condition is directly related to operational safety and energy output. However, FWT vibration signals exhibit significant non-stationary and multi-scale characteristics, with damage-sensitive features of different damage patterns spanning multiple temporal scales. Existing methods fail to sufficiently extract and fuse multi-scale damage-sensitive features. To this end, this paper proposes a novel Residual Dual Attention Multiscale Network (RDAMNet). The network innovatively designs a signal-level multi-scale decoupling strategy that extracts damage-sensitive features at different scales from complementary signal representations through a multi-branch differentiated architecture. Furthermore, an ECA-SE dual attention mechanism is designed to collaboratively enhance damage-related channel responses at both the feature extraction and fusion stages. Multiple independent experimental results on a publicly available dataset demonstrate that RDAMNet achieves a mean damage recognition accuracy and a weighted F1-score of 95.39% and 95.37%, respectively, significantly outperforming five compared methods. Cross-condition generalization experiments further demonstrate that RDAMNet maintains mean accuracies exceeding 94% across different wind speed and wind direction combinations, validating its stability across operating conditions. Moreover, RDAMNet only contains 663,783 parameters with a single-sample GPU inference time of 5.35 ms, exhibiting a favorable performance–efficiency trade-off. The ablation study verifies the effective contribution of each core component, and branch importance analysis, together with Grad-CAM visualization, further substantiates the multi-scale feature learning capability of the network. The proposed method provides an effective technical approach for intelligent structural health monitoring of FWTs in complex oceanic environments. Full article

Microseisms, as the most energetic component of the Earth’s background noise field, represent a forefront area of research where precise location of their sources is paramount. This study systematically investigates the spatiotemporal characteristics of 50 mHz Rayleigh wave microseisms using the dense Shandong array deployed in eastern China, through beamforming and a multi-array combined analysis. The results reveal that the incident direction of the Rayleigh waves exhibits distinct temporal and seasonal variations, primarily originating from four back-azimuth sectors. To further constrain the source regions, we integrate background noise data from the Alaska array and the Venezuela array (supplemented by the Indonesia array). The multi-array product back-projection, by cross-constraining back-azimuths from geographically separated arrays, mitigates the inherent ambiguity of single-array analyses and enables robust global source localization. This approach not only improves the reliability of source attribution but also demonstrates the potential of using microseismic noise as a passive tool for monitoring ocean wave activity and investigating solid-Earth structure. The combined analysis identifies four microseism source regions (M1–M4): the Bering Sea–Gulf of Alaska–Aleutian Islands, the central South Pacific, the southwestern Indian Ocean off southern Africa, and the northeastern North Atlantic–Northern Europe. These source regions fundamentally correspond to areas of elevated significant wave height, confirming the coupled ocean–solid Earth excitation mechanism. These findings provide a methodological basis for future applications of multi-array microseismic monitoring in ocean-climate studies and seismic imaging. Full article

Background and Aims: Rapid digital transformation has generated pervasive attentional disruption in research and professional settings, raising the question of how the temporal conditions that support deep scientific thinking can be preserved. Our narrative review aimed to (i) synthesize neurobiological evidence on the mechanisms through which task-irrelevant digital interruption impairs deep thinking; (ii) discuss the conditions required for deep thinking and the potential threats posed by contemporary developments, including generative artificial intelligence-related cognitive offloading; and (iii) elaborate evidence-based, multi-level recommendations for research institutions. Methods: Targeted searches of PubMed, Google Scholar, and Web of Science (January 2010–September 2025) were conducted using terms spanning attentional neuroscience, digital distraction, neuroplasticity, and cognitive performance, supplemented by forward and backward citation tracking. Peer-reviewed empirical studies, meta-analyses, and theoretical frameworks addressing neurobiological mechanisms of sustained attention and the cognitive effects of digital interruption in professional and/or research settings were included. Results and Interpretation: Deep thinking and protected thinking time are treated as distinct constructs: the former as a sustained, integrative cognitive process supported by coordinated executive control and default mode network activity, the latter as uninterrupted temporal intervals within which that process can occur. Repeated engagement with task-irrelevant digital stimuli is associated with cortico-striatal strengthening and prefrontal-parietal under-consolidation, producing a plasticity paradox in which attentional fragmentation becomes self-reinforcing. The emergence of generative artificial intelligence introduces a qualitatively distinct threat through voluntary cognitive offloading, which reduces deep engagement independently of attentional distraction. Conclusions: Evidence-based strategies spanning individual, team, organizational, technological, and assessment levels are available to preserve protected thinking time. Direct evidence linking these intervals to specific research-impact outcomes remains limited, and institutional interventions should be prospectively evaluated. Full article

Recurring spacecraft anomalies may reflect anomaly-prone system states shaped jointly by spacecraft geometry, recent anomaly history, and space weather forcing, rather than isolated responses to single external drivers. Distinguishing these contributions is important for interpretable anomaly monitoring in geostationary orbit. This study develops a satellite-stratified case–control framework for GOES-16/17 Extreme Ultraviolet and X-ray Irradiance Sensors (EXIS) Space Wire (SpW) anomaly records. Orbital–illumination descriptors, same-satellite event history variables, and space weather variables from the NASA OMNI database are integrated within chronological train–test validation, supported by null-baseline comparison, stratified bootstrap confidence intervals, exclusion window sensitivity analysis, feature group ablation, cross-satellite testing, and calibration diagnostics. Orbital–illumination variables provide weak but reproducible discrimination, event history descriptors capture temporal clustering, and space weather variables add complementary held-out information. The full Space Environment-Integrated Model (SEIM) reached a test area under the receiver operating characteristic curve (AUC) of 0.7540, while a compact train-only L1-selected clean Top-15 model achieved comparable held-out discrimination with a test AUC of 0.7588 after excluding direct near-neighbor history variables. Bootstrap comparisons indicate that this small difference is not statistically significant. Calibration diagnostics further confirm that fitted scores should be interpreted as discriminative anomaly-state indicators under the case–control design rather than as calibrated operational anomaly probabilities. Full article

Information cascades in online social networks arise from stochastic interactions among user behavior, temporal activation, and graph-structured exposure. Early prediction of cascade outcomes remains difficult because only a short diffusion prefix is observable, while future propagation depends on sparse user-level transitions across a heterogeneous social network. This study develops a graph-conditioned stochastic modeling framework for early Twitter cascade prediction. Retweet cascades are formulated as history-dependent stochastic processes over a finite user vocabulary, and a causal dual-head Transformer is used to infer cascade virality and logarithmic final size from short observed prefixes. To incorporate social-network structure, user embeddings pretrained from the follow graph are introduced as external structural priors. A controlled ablation design separates the effects of random embeddings, graph-pretrained embeddings, frozen structural priors, and handcrafted feature fusion. Experiments on Higgs Twitter retweet cascades show that direct full-vocabulary next-user prediction is statistically fragile under sparse short-prefix observations, motivating macro-level cascade outcome prediction. Among the evaluated configurations, the frozen graph-pretrained Transformer achieves the strongest overall balance, reaching an AUC of 0.819, a Brier score of 0.151, and an RMSE of 0.192, while the causal Transformer without a graph prior already surpasses logistic regression and approaches Random Forest; however, gains over competitive baselines are modest and statistically significant only in selected pairwise comparisons. Calibration analysis, bootstrap confidence intervals, and paired statistical tests confirm that graph-derived user priors provide more reliable improvements than sequence modeling alone under short-prefix sparse observations. These findings indicate that graph-conditioned structural priors offer a promising complement to causal sequence modeling for early Twitter cascade prediction. Full article

Full-physics reservoir simulation for CO₂ storage becomes computationally expensive when many operational schedules must be screened, motivating machine-learning (ML) surrogates that reduce simulation burden while preserving the essential physics-driven response. We propose an activity-based partitioning methodology that produces an interpretable applicability map, identifying regions where surrogate substitution is expected to be reliable and regions where highly active dynamics make it unsafe. In this work, we focus exclusively on the slow-varying region and develop proxy models for pressure and saturation time derivatives in that domain. The fast-varying region is intentionally excluded, and no fully coupled hybrid simulator is claimed at this stage. The partition is constructed from temporal changes in derivative signals and aggregated across multiple schedules to obtain a conservative, scenario-robust delineation. For slow cells, local stencil-based neural proxies leverage overlapping time windows and features describing the local state, schedule forcing, and injector influence. Because saturation derivatives in the slow region are strongly zero-inflated, with many cells remaining outside the advancing CO₂ plume for long periods, a two-stage strategy is adopted: first detecting whether meaningful change occurs and then predicting the derivative magnitude only when active, with additional smoothing to suppress near-zero artifacts. The framework also supports selective surrogate deployment over user-selected time windows. The objective is therefore to establish a conservative zone of applicability for derivative-based ML updates, rather than to demonstrate full simulator replacement or end-to-end coupled acceleration. In the case study, 5914 of the 8243 grid blocks evaluated by the proxy workflow were classified as slow-varying, corresponding to 71.7% of the evaluated proxy-analysis domain. For the blind schedule, full-rollout pressure reconstruction produced mean absolute errors of 5.34, 3.69, and 2.80 psi over early, middle, and late time-window groups, respectively. In a future coupled implementation using the same partition, these 5914 cells could be advanced by the ML proxy, while the remaining dynamically active or unsupported cells would remain under full-physics treatment. This would reduce the full-physics active-cell count from 9212 to 3298 in the future coupled setting, although direct wall-clock acceleration remains to be quantified after simulator integration. Full article

Hydraulic fracturing in naturally fractured rock is governed by complex interactions between fluid flow, rock deformation, fracture propagation, and induced seismicity. In this study, a fully coupled hydro-mechanical framework based on the FDEM is developed to investigate fracture evolution and seismic responses during fluid injection in fractured rock masses. Three representative horizontal stress ratios (R = 1.0, 1.5, and 2.0) were considered to investigate the influence of stress anisotropy on fracture propagation and induced seismicity. The results demonstrate that stress anisotropy exerts a dominant control on fracture propagation patterns, fluid pressure diffusion, and induced seismicity. Under low stress ratios, fracture propagation is diffuse and strongly influenced by pre-existing fractures, whereas higher stress ratios promote localized, directional fracture growth controlled primarily by the stress field. Fluid pressure becomes increasingly concentrated with increasing stress ratio, leading to higher injection pressures and more pronounced pressure fluctuations. The spatial and temporal evolution of mean stress and volumetric strain closely follows that of fluid pressure, indicating that fluid pressurization directly controls effective stress reduction and associated deformation. Seismic analysis reveals a systematic decrease in the Gutenberg–Richter b-value with increasing stress ratio, indicating a transition from distributed micro-fracturing to more coherent fracture reactivation and larger seismic events. Under quasi-steady injection pressure conditions, fracture propagation is found to be episodic and unstable, as evidenced by pronounced positive and negative spikes in the fracture volume change rate and associated pressure fluctuations; these are accompanied by intermittent fracture opening and closure, stress redistribution, and temporary reductions in cumulative seismic moment. These findings provide new insights into the coupled mechanisms governing hydrofracturing-induced seismicity and have important implications for the assessment and mitigation of seismic risks in subsurface engineering applications. Full article