Search Results (527)

High-frequency volatility and extreme tail risks in offshore wind power pose severe challenges to grid stability and economic operation. Conventional storage planning often relies on deterministic profiles or static allocation rules, failing to capture the non-stationary temporal dynamics of marine wind resources. To bridge this gap, this paper proposes a closed-loop framework that integrates ultra-short-term probabilistic forecasting with dynamic hybrid energy storage optimization. A novel Dual-Channel Residual Network is developed to provide well-calibrated predictive uncertainty quantification, which explicitly drives a Prediction-Guided Dynamic Hybrid Storage Optimization Framework. By dynamically coordinating lithium-ion batteries and liquid air energy storage based on evidential predictive variance, the proposed approach achieves superior synergy between short-term power response and long-duration energy shifting. Case studies on an offshore wind farm validate that the framework significantly reduces the Levelized Cost of Energy and loss-of-load risks while enhancing frequency regulation capabilities compared to state-of-the-art benchmarks. Full article

Accurate solar radiation estimation is critical for optimizing solar energy applications. This study divided 819 meteorological stations in China into six solar radiation zones using k-means, hierarchical, and bisecting k-means clustering based on daily relative sunshine duration. Correlation analysis and feature importance evaluation were conducted to quantify the contributions of key meteorological variables. A comparison of models considering regional heterogeneity was performed. Six sunshine-based empirical models, three machine learning models (Random Forest, Support Vector Machine, and Extreme Gradient Boosting), and two deep learning models (Long Short-Term Memory and Gated Recurrent Unit) were systematically evaluated across 98 stations with observed solar radiation data. Model performance was assessed using the coefficient of determination (R²), mean absolute error (MAE), root mean square error (RMSE), and normalized RMSE (NRMSE). Results showed that k-means clustering outperformed the other two methods and was adopted for final zoning. The correlation analysis identified sunshine duration (S), extraterrestrial radiation (R_a), temperature difference (ΔT), and maximum temperature (T_max) as the dominant influencing factors, with clear regional heterogeneity. The deep learning models, particularly LSTM (R² = 0.939, RMSE = 1.702 MJ/m/²/d¹, MAE = 1.319 MJ/m/²/d¹, NRMSE = 0.046), achieved the highest accuracy, followed by GRU, XGB, SVM, and RF. Among the empirical models, Model 5 performed best in Zones 1, 3, 4, and 5, while Model 6 was optimal in Zones 2 and 6. The key novelty of the study is an integrated zoning–prediction framework for regional solar radiation estimation, combining clustering validation, correlation analysis, empirical model calibration, and deep learning benchmarking, with enhanced physical interpretability and prediction accuracy. Full article

Light Detection and Ranging (LiDAR) three-dimensional (3D) object detection degrades under point sparsity, outliers, coordinate noise, and calibration drift, yet detector evaluation remains largely limited to clean benchmarks. This study focuses on sensing robustness rather than detector redesign. We introduce Bounded Graph Conditioning (BGC)—a deterministic pre-voxelization front-end that applies k-nearest-neighbor (kNN) neighborhood averaging with bounded residual correction upstream of an unchanged detector backbone. BGC is evaluated together with a reproducible sensor-degradation stress protocol and a risk-constrained operating-boundary analysis. Experiments on KITTI with PointPillars, SECOND, and Voxel R-CNN show that BGC most clearly improves retained detection quality and feasible operating coverage under strong noise and strong outlier stress; gains under other degradation types are smaller and backbone-dependent. In the primary score-level box-disjoint calibration/test evaluation on SECOND, maximum feasible coverage at a target risk bound of $0.2$ improves from $0.0754$ to $0.1374$ under strong noise ($\sigma =0.10$ m) and from $0.1323$ to $0.1591$ under strong outliers ($p=0.10$); a cross-backbone check on Voxel R-CNN confirms the same direction ($0.1860\to 0.2864$). Comparison with traditional filtering (SOR and ROR) reveals complementary strengths across fault types. A range-adaptive BGC variant that adjusts parameters per distance bin further improves performance under mixed unknown faults, spherical-coordinate noise, and on a dataset-matched nuScenes validation (adaptive BGC mAP/NDS: $0.2687/0.4493$ vs. baseline $0.2471/0.3846$ under strong noise). Severe translation drift collapses all configurations to full rejection, exposing an explicit sensing boundary beyond the reach of local conditioning. These results support BGC as a practical sensor-side robustness enhancement under the studied degradation protocol, with conditional rather than universal applicability across backbones and fault types. Full article

(1) Background: Advances in machine learning (ML)-based survival modeling enable the analysis of high-dimensional biomedical data. However, many approaches rely on the proportional hazards (PH) assumption, which is frequently violated in oncology and can limit the interpretability of hazard ratio-based results. Using Estrogen Receptor (ER) status in the METABRIC breast cancer cohort as a case study, we propose a framework that integrates machine learning survival models with Restricted Mean Survival Time (RMST) to provide a more robust and clinically interpretable approach for survival analysis under non-proportional hazards. (2) Methods: Overall survival was analyzed in 1104 patients. PH violations were confirmed using Schoenfeld residuals and Kaplan–Meier inspection. We compared four models: stratified Cox Elastic Net (Cox E-Net), Random Survival Forest (RSF), Gradient Boosting Survival Analysis (GBSA), and DeepHit. Performance was assessed using Harrell’s C-index, time-dependent IPCW C-index, and Integrated Brier Score (IBS). RMST at 180 months was utilized to quantify absolute survival differences between ER subgroups. To improve the stability of the estimates, 200 bootstrap resamples were performed, and 95% confidence intervals were derived from the bootstrap distribution. (3) ER status demonstrated significant PH violation (p < 0.005) with crossing survival curves. Discrimination (C-index 0.664–0.725) and calibration (IBS 0.149–0.169) were comparable across models, with RSF achieving the highest overall performance. Despite similar accuracy, survival curve structures differed substantially. Cox E-Net and RSF reproduced the observed crossing pattern, whereas GBSA generated smoother trajectories and DeepHit showed marked compression of subgroup separation. In the independent test cohort, the empirical RMST difference at 180 months was 16.6 months (ER-positive: 130.4; ER-negative: 113.8). Model-based RMST differences ranged from 1 month (DeepHit) to 27 months (Cox E-Net), with RSF and GBSA (12.8 and 13.8 months) most closely approximating the empirical benchmark. (4) Conclusions: We propose a novel, model-agnostic ML + RMST framework that addresses non-proportional hazards while providing quantifiable, time-specific clinical benefit. Moreover, models with similar discrimination and calibration produced markedly different survival curve behavior and absolute RMST estimates, demonstrating that accuracy metrics alone are insufficient for clinical interpretation. By linking prognostic modeling with absolute survival quantification, this framework advances survival evaluation beyond relative risk ranking toward individualized, clinically meaningful decision support. Full article

Accurate and trustworthy prediction of lithium-ion battery aging remains challenging due to multi-mechanistic degradation, cell-to-cell variability, and distribution shift between laboratory calibration and deployment. Fractional-order models have been proposed to capture long-memory effects in electrochemical systems; however, it remains unclear when such memory is empirically identifiable and beneficial within the common prognostics abstraction of state-of-health (SOH) versus cycle index. This work develops a fully reproducible computational pipeline for mechanistic battery aging based on a Caputo fractional differential equation (FDE) and evaluates its cross-cell generalization on open NASA cycling data. Parameters are identified using bounded robust nonlinear least squares and validated under a strict transfer protocol: calibration on cells B0005/B0006 and evaluation on held-out cells B0007/B0018 without refitting. The fractional model is benchmarked against a classical ODE surrogate, an ECM-inspired resistance-proxy baseline, and one-step-ahead machine-learning predictors. Uncertainty quantification is performed via parameter bootstrap and subsequently calibrated using conformal correction to target nominal coverage under transfer. Results show that the fractional order tends to collapse toward the integer-order limit (α → 1) in this dataset, indicating limited evidence of additional long-memory at the SOH-versus-cycle level under the considered protocol, while robust identification remains essential for stability. Calibrated prediction intervals achieve near-nominal coverage on held-out cells, highlighting the importance of UQ calibration under cell-to-cell shift. The proposed scripts and environment specifications enable direct replication and facilitate future extensions to stress-aware fractional models and hybrid physics–ML approaches. Full article

The concentration of tourist flows along short urban links caused by cruise stops in medium-sized Mediterranean ports exacerbates traffic and localized environmental externalities. This study evaluates the willingness to pay (WTP) of cruise passengers for an electric tram that would connect the Gaženica Port with Zadar’s historic center, an intervention designed to cut travel time and reduce on-street congestion and emissions. Over the course of two seasons, a two-wave, two-site, in-person survey was conducted at the port and in the city center. The instrument adopts a double-bounded dichotomous choice (DBDC) contingent valuation design with randomized starting bids that were calibrated using a pre-test that benchmarked prevailing transport pricing. Primary WTP estimates are obtained from a binary choice model with socio-demographic and environmental covariates; whereby inference relies on cluster-robust errors. Robustness is assessed through three complementary checks that do not require additional data: (i) a bivariate specification to account for within-respondent correlation between first and follow-up bids; (ii) Turnbull nonparametric bounds for the interval-censored WTP distribution; and (iii) starting-point tests using split-sample estimation and bid-set indicators. A spike adjustment based on “no–no at the lowest bid” responses is explored where appropriate. Beyond its methodological contribution, this research advances the sustainable tourism development discourse by quantifying visitors’ financial support for low-emission urban mobility infrastructure that mitigates environmental stresses while preserving residential life quality. The results integrate cruise tourist management with the more general goals of resilient and sustainable urban destinations by offering a decision-ready value input for port-city mobility planning in historic Mediterranean centers. Full article

Human serum albumin (HSA) binding critically influences drug distribution and pharmacokinetics. In this study, HSA affinity chromatography was integrated with machine-learning-based quantitative structure–retention relationship (QSRR) modeling to elucidate structural determinants of albumin binding in a library of 115 fluoroquinolone (FQs) derivatives. Experimentally determined logk_HSA values were obtained using biomimetic chromatography, and these were then used as modelling endpoints. Following descriptor reduction via Least Absolute Shrinkage and Selection Operator (LASSO) and systematic benchmarking of 42 regression algorithms, support vector regression (SVR) and nu-support vector regression (ν-SVR) with radial basis function kernels demonstrated superior predictive performance. A parsimonious 12-descriptor ν-SVR model achieved strong calibration and validation metrics (R² = 0.916, Q²_test = 0.823, concordance correlation coefficient (CCC) = 0.899) and satisfied Organisation for Economic Co-operation and Development (OECD) criteria, including applicability domain assessment. Shapley Additive exPlanations (SHAP)-based interpretation revealed that albumin binding is governed by a balance between hydrophobic surface area and distributed electronic properties, whereas excessive localized polarity and quaternary ammonium functionalities reduce affinity. This experimentally anchored and interpretable modeling framework provides mechanistic insight into HSA binding in fluoroquinolones and offers a robust tool for rational pharmacokinetic optimization. Furthermore, in order to make the model easily accessible to users, we have packaged it in the form of an online application. Full article

Multi-campus AI data centers procure identical hardware and service SKUs across geographically heterogeneous locations, yet finance and operations require a single system-level benchmark (“world price”) per SKU for budgeting, chargeback, and capacity planning. Naive deployment-weighted aggregation preserves total cost but can induce Simpson-type aggregation bias, where heterogeneous location mixes reverse global SKU rankings and weaken managerial decision signals. This study formalizes the problem of location-robust, cost-preserving aggregation and develops two mathematically structured operators for production cost pipelines. The first operator applies a two-way fixed-effects decomposition to separate global SKU effects from campus-specific premia, followed by normalization to guarantee exact cost preservation. This yields an interpretable benchmark that performs well when campus coverage is sufficiently broad and location effects remain approximately additive. The second operator solves a constrained convex common-weight optimization, producing a unified set of non-negative campus weights that preserves total cost while providing the strongest protection against dominance reversals in the ordered setting. Simulation experiments and a semi-real calibrated AI datacenter OPEX illustration show that both operators substantially improve ranking stability relative to naive blending, while the convex operator serves as the more conservative safeguard under adverse heterogeneity. The resulting detect–correct–validate workflow provides a scalable decision-support framework for robust cost aggregation in distributed AI infrastructure and illustrates how symmetry-preserving aggregation operators can stabilize benchmarking in large heterogeneous systems. Full article

Reliable short-term air traffic forecasts are important for operational planning in national airspace systems. This study develops a transparent forecasting framework for Türkiye’s monthly aircraft movements using publicly available data from the General Directorate of State Airports Authority (DHMİ) for 2018–2025. Because DHMİ releases may follow cumulative within-year reporting, month-specific increments are reconstructed through within-year differencing and checked through simple audit procedures. The empirical analysis compares seasonal naïve, ETS, and a constrained SARIMA family under leakage-free evaluation, combining a strict 2025 holdout with expanding-window rolling-origin validation. Forecast performance is assessed using standard accuracy metrics and complemented by Diebold–Mariano comparisons, which are interpreted cautiously, given the short holdout length. To examine instability around the pandemic period, this study also reports structural-break and stability diagnostics as supportive evidence rather than definitive identification. Uncertainty is evaluated through backtested 80% and 95% prediction intervals, comparing nominal SARIMA intervals, parametric bootstrap, split conformal prediction, and adaptive conformal inference (ACI). The results show that SARIMA provides the strongest point-forecast performance among the benchmarked models, while adaptive conformal calibration offers a useful balance between empirical coverage and interval width under changing conditions. Overall, this study provides a reproducible and operationally interpretable baseline for national air traffic forecasting in Türkiye and a clear benchmark for future multivariate extensions. Full article

Satellite-derived bathymetry (SDB) offers a practical alternative for mapping shallow reefs in remote oceanic settings where acoustic surveys are costly and logistically constrained. Here we benchmark PlanetScope 8-band (3 m) surface reflectance—an underused commercial constellation for reef SDB—using ICESat-2 Advanced Topographic Laser Altimeter System (ATLAS) ATL03 photon data (Release 006) as independent vertical control. Seventeen ATL03 ground tracks (2019–2025) were processed using geometric filtering, photon classification, and explicit air–water refraction correction. This yielded 5171 candidate seafloor observations, of which 5021 were co-located with valid PlanetScope water pixels after Usable Data Mask screening (UDM2/UDM2.1), sun-glint correction, and reflectance quality screening. Four SDB formulations (Lyzenga, Bierwirth, and Stumpf) were calibrated and independently validated using depth-stratified train/validation partitions (70/30, 80/20, and 90/10). Across partitions, the multiband polynomial model of Lyzenga 2006 generalized best (R² = 0.843–0.859; RMSE = 1.734–1.813 m; bias = −0.070 to −0.081 m), followed by Bierwirth (R² = 0.826–0.845; RMSE = 1.818–1.904 m). Lyzenga 1985 reported lower skill (RMSE ≈ 3.1 m), while the Stumpf log-ratio failed in independent validation. ICESat-2 photon bathymetry provides repeatable point-based control in clear waters but remains less precise than echo sounding due to photon classification and spatial-support effects; therefore, uncertainties and applicability limits must be reported. Overall, PlanetScope 3 m, 8-band surface reflectance supports reproducible reef-scale SDB in Seaflower under the evaluated conditions, with Lyzenga 2006 as a robust baseline. Full article

Nigeria entered the 2020 COVID-19-related oil price downturn without the fiscal buffers that numerous resource-rich economies had built over time. Despite heavy dependence on petroleum revenues, the country has made limited use of stabilization tools such as structured hedging programs, sovereign savings mechanisms, or strategic reserves, leaving public finances exposed to external shocks. Drawing on political choice theory and the resource governance literature, this study examines how institutional conditions shaped crisis management during the 2020 oil price collapse and the COVID-19 pandemic. The study combines qualitative institutional analysis with a stochastic counterfactual simulation. It compares Nigeria’s policy approach with those of oil-producing countries including Mexico, Saudi Arabia, the United Arab Emirates, Angola, and Ghana, using data from the IMF, World Bank, Afreximbank, and peer-reviewed sources. The counterfactual simulation is calibrated to Nigeria’s 2019 federal budget oil benchmark of US $60 per barrel, with the IMF’s 2019 petroleum price assumption used as a robustness check. The model treats hedging as a form of partial fiscal insurance rather than full stabilization. Results suggest that hedging sufficient to offset 10%, 20%, and 30% of the shock would have improved 2020 GDP decline from −1.80% to approximately −1.62%, −1.44%, and −1.26%, respectively. The analysis identifies institutional gaps in Nigeria’s use of hedging, sovereign savings, and reserve infrastructure. The counterfactual results indicate that even modest oil hedging could have meaningfully softened the 2020 downturn, with the 20% scenario reducing GDP contraction by an estimated 0.36 percentage points. These findings suggest that governance constraints contributed materially to fiscal vulnerability. The study proposes a four-pillar framework centered on risk hedging, revenue savings, strategic investment, and institutional reform to strengthen fiscal stability and resilience to external shocks. Full article

Three-Dimensional (3D) face reconstruction from monocular Red-Green-Blue (RGB) imagery remains a fundamental yet ill-posed challenge in computer vision, with applications in biometrics, augmented reality/virtual reality (AR/VR), and intelligent visual sensing systems. While deep learning has significantly improved reconstruction fidelity and realism, existing surveys primarily focus on network architectures in isolation, often overlooking how sensing conditions, data acquisition protocols, and geometric calibration influence reconstruction reliability and evaluation outcomes. This paper presents a sensor-aware, end-to-end review of deep learning-based 3D face reconstruction and introduces a unified modular framework that connects sensing hardware, data acquisition, calibration, representation learning, and geometric refinement within a coherent pipeline. The reconstruction process is organized into four stages: sensor-driven acquisition and calibration, landmark estimation and feature extraction, 3D representation and parameter regression, and iterative refinement via differentiable rendering. Within this framework, we examine how sensor characteristics, calibration accuracy, representation models, and supervision strategies affect reconstruction accuracy, perceptual quality, robustness, and computational efficiency. We further synthesize the reported results across widely used benchmarks using both geometric and perceptual metrics, highlighting trade-offs between reconstruction fidelity and deployment constraints. By integrating sensing-aware analysis with architectural evaluation, this survey provides practical insights for developing scalable and reliable 3D face reconstruction systems under real-world conditions. Full article

Robotics is increasingly positioned as an enabling technology for precision agriculture, where management actions must be spatially and temporally targeted under constraints on labour, input use, safety, and environmental impact. This review synthesises studies on agricultural field robotics and organises the literature along four complementary axes: task (monitoring, weeding, spraying, and harvesting), platform (UGV, UAV, gantry/fixed-structure, greenhouse robot, and hybrid systems), autonomy-stack module (perception, localisation, planning, control, actuation, safety, and human–robot interaction), and evaluation setting (lab, greenhouse, open-field single season, and open-field multi-season/multi-site). Across these dimensions, this review analyses how platform constraints shape sensing geometry, actuation capability, localisation reliability, energy/endurance, supervision burden, and safety requirements. It further examines enabling technologies that recur across tasks, including vision and multimodal perception under occlusion and illumination variability, localisation and mapping under weak or denied GNSS, uncertainty-aware planning in deformable and partially observed environments, and compliant end-effectors for contact-rich operations. Beyond cataloguing systems, this paper emphasises evaluation practice by synthesising core task-relevant metrics, comparing laboratory and field validation settings, and proposing a reporting checklist and benchmark ladder to improve reproducibility and cross-study comparability. This review identifies recurring bottlenecks in domain shift, long-term autonomy, calibration robustness, crop-safe actuation, and safety assurance near humans, and it concludes with a staged research roadmap linking near-term evaluation reform to longer-term credible multi-site autonomy. Overall, this paper provides a structured framework for interpreting agricultural robotic systems not only by application but also by deployment context, system maturity, and evaluation credibility. Full article

A robust analysis of water use in major food production systems is crucial for improving their productivity and sustainability in water-scarce arid and semi-arid regions like Punjab (India) facing the depletion of groundwater resources. This study aimed to assess blue water use and blue water productivity in dairy farming systems across different farm sizes in Punjab. Comprehensive monitoring and assessment of water use over a full year (from July 2022 to June 2023) was conducted on 24 dairy farm sheds in Punjab, revealing significant variability in their blue water use (measured in L per adult animal per day) and blue water productivity quantified as kg of fat- and protein-corrected milk (FPCM) produced per m³ of the blue water consumed. The variability was influenced by factors such as livestock species, farm size (medium with 15–25 livestock, large with 25–100 livestock, and commercial with >100 livestock), bathing and servicing routines, and energy use patterns. The average dairy livestock total blue water consumption varied from 112 ± 14 to 131 ± 19 L per adult animal per day, with 20–40% higher livestock drinking water and about six times higher livestock bathing and serving water used during the summer months. Interestingly, a large share (45%) of the average total blue water consumption is contributed by indirect water consumption via the use of energy (electricity and diesel) in dairy farm sheds. Dairy milk blue water productivity was quantified higher, ranging from 154 ± 11 to 225 ± 59 kg FPCM per m³ in buffalo- and crossbred cattle-based dairy farm sheds. However, indigenous cattle showed a lower blue water productivity ranging from 56 to 97 kg FPCM per m³, reflecting their lower milk yields and limited use of intensified management practices. The state-level water scarcity footprint (WSF) of Punjab dairy farm sheds was quantified at 4870 million m³ world-eq, which showed a significant spatial variation among Punjab districts. However, the results of this study offer novel seasonally and spatially disaggregated benchmarks of blue water consumption, blue water productivity, and the water scarcity footprint of Punjab’s dairy farming sheds. This new information is crucial for the development of locally calibrated and validated models for improving the water productivity and sustainability of dairy farming across Punjab and other similar arid and semi-arid regions in Southeast Asian countries. Full article

Applying large language models (LLMs) to industrial fault classification is hindered by the mismatch between tabular sensor data and text-based inputs and by the high memory cost of fine-tuning billion-parameter models on edge hardware. This paper presents TabEng-QLoRA, a framework with three contributions: (1) a criticality-aware serialization module that converts tabular sensor records into structured prompts, placing fault-critical features in semantically prominent positions; (2) a saliency-guided rank allocation mechanism that profiles layer-wise activation norms on a 500-sample calibration set and assigns adapter ranks in three tiers (r ∈ {8, 16, 32}); and (3) a feed-forward domain router for automatic adapter selection (98.1% accuracy, 0.6 ms latency). Experiments on three public benchmarks (the AI4I Predictive Maintenance Dataset) using three foundation models (LLaMA-3-8B, Mistral-7B, and Qwen2-7B) show that TabEng-QLoRA achieves a mean macro F1 of 0.908, a 10.6% gain over standard QLoRA, within 4.6–5.2 GB peak GPU memory. The framework closes 82% of the gap to full fine-tuning, while offering advantages in cross-equipment transfer learning (zero-shot macro F1: 0.743 vs. 0.341 for XGBoost retrained on 20% of target-domain data, as XGBoost cannot perform zero-shot transfer). Ablation results confirm statistically significant contributions from all three components (p < 0.001). Full article