Search Results (791)

This paper proposes an annular suction caisson specifically designed to reinforce in-service monopiles and upgrade existing offshore wind farms to accommodate larger-capacity wind turbines. During penetration of the annular suction caisson into clay, the existing monopile restricts the inward migration of soil into the internal space of the caisson, promoting upward soil displacement and consequently increasing the height of the soil plug formed inside the caisson. In addition, the strain-softening behavior causes varying degrees of strength degradation in the clay along the caisson wall. The softened zones extend approximately one caisson wall thickness on the inner side and 1.2 times the wall thickness on the outer side of the caisson. Both effects should be considered for accurately predicting the penetration resistance of annular suction caissons. Therefore, three-dimensional large-deformation finite element analyses were performed to investigate the penetration behavior of annular suction caissons in strain-softening clay. A comprehensive parametric study was conducted to quantify the soil plug heave and overall penetration resistance. Meanwhile, the soil flow mechanism at the caisson tip, the evolution of clay strength along the caisson wall, and the formation characteristics of the internal soil plug were systematically examined. Based on the numerical results, a theoretical approach was developed to evaluate the penetration resistance of annular suction caissons. Full article

Static wake steering can increase wind-farm power production, but its application to floating offshore wind farms requires assessment of the coupled wake, platform, structural, and station-keeping response. This study evaluates whether power-maximizing static yaw setpoints identified using the steady, control-oriented FLORIS model retain their benefit when transferred without re-optimization to a coupled FAST.Farm floating wind-farm model. The reference farm comprises four IEA Wind 15 MW turbines mounted on VolturnUS-S semi-submersible platforms. Greedy and static wake-steering operations are compared at three below-rated wind speeds, three sea states, and five matched turbulent-inflow realizations, resulting in 90 farm-level FAST.Farm simulations. Wake behavior is characterized through wake-center deflection, meandering, and velocity-deficit profiles, while turbine and mooring fatigue responses are evaluated using paired damage-equivalent-load statistics. Static wake steering increases mean farm power under all nine investigated wind–wave conditions. The gains are approximately 5.1–$5.2\%$ at $7\mathrm{m}{\mathrm{s}}^{-1}$, $5.0$–$5.1\%$ at $8\mathrm{m}{\mathrm{s}}^{-1}$, and $4.0$–$4.2\%$ at $9\mathrm{m}{\mathrm{s}}^{-1}$, with all paired $95\%$ confidence intervals remaining above zero. The gain results from a power redistribution in which the intentionally yawed upstream turbine incurs a local loss that is exceeded by the combined recovery of the downstream turbines. The fatigue response is strongly component- and turbine-dependent. The paired farm-mean blade-root DEL decreases by $0.82$–$2.24\%$, whereas the tower-base DEL increases by $0.76$–$2.78\%$, and the FairTen1 response generally increases by $0.88$–$2.92\%$. The farm-mean yaw-bearing response is mixed, ranging from a $1.15\%$ reduction to a $4.32\%$ increase. Turbine-level analysis reveals larger localized penalties, reaching approximately $10.4\%$ for the yaw-bearing DEL and $12.8\%$ for FairTen1. Spectral analysis associates the yaw-bearing response with yaw-induced aerodynamic and structural excitation, while the tower-base response is strongly influenced by low-frequency wave–platform dynamics. A complementary FLORIS sensitivity analysis demonstrates that the optimized aerodynamic benefit depends strongly on wind direction, spacing, wind speed, and turbulence intensity. For a Tampen-derived 11-turbine layout, resource weighting over the modeled 4–$13\mathrm{m}{\mathrm{s}}^{-1}$ interval produces an annual energy-contribution increase of $3.653\mathrm{GWh}{\mathrm{year}}^{-1}$, or $0.921\%$. These results provide numerical evidence that static wake steering can retain a positive power benefit in a coupled floating wind-farm environment, but controller assessment must include turbine- and component-specific dynamic loads rather than farm power alone. Full article

Landscape Character Assessment (LCA) has struggled with scale mismatches and inadequate treatment of cultural expression when applied to high-density cultural landscapes in Asia. On the Chengdu Plain, Western Sichuan, China, remote sensing shows convergent settlement patterns across case sites, but fieldwork reveals substantial divergences in landscape character. We examined five cases representing different rural development models in the Chengdu metropolitan area. By integrating 10 m Sentinel-2 imagery and with systematic fieldwork (3200 photographs and 42 semi-structured interviews, July–August 2023), we constructed a framework that combines macro-scale spatial analysis with micro-scale field investigation. For macro-level quantification, six landscape pattern indices were adopted: number of patches (NP), patch density (PD), contagion index (CONTAG), aggregation index (AI), Shannon’s diversity index (SHDI), and Shannon’s evenness index (SHEI). The micro-level investigation followed a three-tier scheme (natural environment, settlement space, architectural details) to record vertical ecological communities, spatial expressions of cultural embeddedness, and visual landscape character. Indicator screening employed the KJ method and two rounds of Delphi consultation with ten experts; weights were then determined through the Analytic Hierarchy Process (AHP). Macro-level results display a clustered, low-fragmentation pattern (woodland AI = 97.48, water body AI = 99.40), anchored by a stable mosaic of Linpan, farmland, water bodies, and homesteads. Micro-level features include multi-strata vertical communities of trees, shrubs, grasses, and water, along with cultural embeddedness expressed through irrigation systems, farming patterns, and courtyard layouts. The VLEAS (Vernacular Landscape Elements Assessment System) framework encompasses three dimensions (natural, cultural, and visual) and comprises 12 indicators. The cultural dimension receives the highest weight (0.42), followed by natural (0.32) and visual (0.26) dimensions. Comparative analysis shows that the framework differentiates cases with similar macro-scale patterns but divergent micro-features, capturing differences that a purely macro-index benchmark misses, and it also reveals trade-offs among ecological integrity, cultural heritage, and visual quality. Sensitivity analysis displays that case rankings remain stable across four extreme weighting scenarios (cultural, ecological, visual, equal), with only one swap (Xingfu Pastoral Park and Nongke Village) under ecological priority and no rank change greater than one position, confirming that the results are not driven by any specific weight assignment. This approach extends the utility of LCA in high-density rural cultural landscapes and offers a practical tool for differentiated conservation and management, especially in tourism-influenced rural settings on the Chengdu Plain. Transferability to other regions would require recalibration of locally relevant indicators. Full article

The uniformity of maize plant spacing serves as a critical indicator for assessing sowing quality, seed vigor, and field seedling emergence stability. However, manual measurement is inefficient, and complex field conditions make automatic seedling detection and plant spacing measurement challenging. Aiming at the challenges of missed detection, insufficient accuracy for small targets, and large errors in plant spacing calculation under complex field conditions, this study constructs a high-quality dataset containing 693 maize seedling images and implements preprocessing enhancement for images degraded by haze or dust. An intelligent maize seedling detection and plant spacing measurement method based on improved YOLOv8 is proposed. The Global Attention Mechanism (GAM) is embedded into the backbone network to strengthen cross-dimension information interaction between channels and spaces, suppress background interference, and reduce the missed detection rate. The Bi-directional Feature Pyramid Network (BiFPN) is adopted to replace the original PAFPN for enhanced multi-scale feature fusion and deep semantic representation. A new 160 × 160 high-resolution small-object detection layer is added to significantly improve the detection performance of weak and small seedlings. Experimental results demonstrate that the improved model achieves a precision, recall, mAP50, and mAP50-95 of 89.4%, 90.3%, 94.4%, and 49.4%, respectively, which are 2.6, 0.5, 1.5, and 2.7 percentage points higher than those of the original YOLOv8 model. These results indicate that the proposed model improved maize seedling detection performance under complex field conditions. Automatic plant spacing calculation is realized based on detection outputs; the average plant spacing of the dataset is 30.42 cm, with a relative error of only 4.93% compared with the preset sowing spacing of 32 cm. The proposed method can efficiently accomplish field seedling identification, plant spacing quantification, and sowing quality evaluation, providing reliable technical support for precision maize sowing, seeder parameter optimization, and intelligent field management, which is of great significance for promoting the intelligent upgrading of grain crop production. Full article

High-pathogenicity avian influenza (HPAI) remains a recurrent threat to poultry production in the Republic of Korea, but production-type-specific clustering has rarely been assessed against the underlying farm population. This retrospective unmatched case–control point-pattern study examined spatial and space–time clustering across six epidemic seasons from 2020 to 2026. We analyzed 486 reported poultry-farm outbreak records together with a national registry of 66,499 poultry farms. The principal analyses compared 204 layer and 204 duck outbreak events with farms of the same production type using the Cuzick–Edwards k-nearest-neighbor statistic, an observed-to-expected clustering-intensity ratio, kernel relative-risk mapping, and spatial and space–time scan statistics. At k = 5, the observed-to-expected intensity ratio was 2.51 for layer events and 1.63 for duck events; this between-type contrast was descriptive because no direct inferential test was performed. Layer-event clustering remained elevated across all seasons, although the extreme 2023–2024 estimate was based on only 15 events. Duck-event clustering was more intermittent and increased in 2025–2026. Premises-level scan statistics identified a broad layer cluster with a relative risk of 3.12 and a smaller duck cluster with a higher local relative risk but far fewer farms. A layer space–time cluster occurred in 2024–2025. These findings support production-type-specific, density-aware surveillance. They do not establish farm-to-farm transmission or implicate egg-handling infrastructure, and they should be interpreted in light of the static registry, repeated-event design, and unmeasured flock size, movements, biosecurity, environmental exposure, and viral lineage. Full article

Public cattle re-identification benchmarks remain limited in scale and rarely provide leakage-controlled open-set and cross-camera evaluation. Using 25,209 curated crops of 197 Holstein cows recorded by four synchronized side-view cameras in a passageway of a working farm, we investigate viewpoint-related shifts in the embedding space and evaluate cattle re-identification under camera-disjoint gallery/query construction. The extractor combines an animal-pretrained MegaDescriptor Swin-B backbone with multi-level projections, learned gating, quality weighting, and two-level aggregation. Under the strictly inductive identity-disjoint cross-camera protocol without camera centering, the proposed model achieved Rank-1 of 55.7% (95% CI: 47.7–63.8%), Rank-5 of 90.0% (86.3–93.7%), and mAP of 70.2% (66.3–74.1%). Under the same protocol, TransReID ViT-B/16 achieved 38.6%, 77.9%, and 55.9%, while OSNet-x1.0 achieved 12.9%, 47.9%, and 30.7%, respectively. The comparison with the pure-GAP configuration was not statistically resolved (Holm-adjusted p = 0.91). In held-out open-set evaluation, the validation-selected global cosine threshold obtained the strongest mean AUROC, F1, and AUOSCR (63.2%, 66.9%, and 44.4%); the corresponding results were 60.2%, 64.4%, and 38.8% for Mahalanobis rejection and 56.9%, 63.2%, and 41.7% for EVT. Zero-shot evaluation at a second corridor on the same farm yielded Rank-1 of 26.6%, Rank-5 of 41.9%, and mAP of 35.3%. These results establish cross-camera feasibility under the defined protocols and same-farm cross-installation transfer, but do not establish cross-farm, cross-device, seasonal, longitudinal, or full production robustness. Full article

Vertical farming (VF) offers a smart way to grow crops in stacked layers inside controlled indoor environments. By doing so, it uses far less land and water than traditional open-field agriculture, making it a promising solution for cities with limited space and resources. In recent years, artificial intelligence (AI), machine learning (ML), and Internet of Things (IoT) technologies have begun to transform vertical farming. These tools are moving the industry away from rigid, rule-based systems toward more flexible, data-driven operations that can adapt in real time. This paper presents a systematic review of 208 peer-reviewed studies from 2015 to 2025. It explores how AI, ML, and IoT are applied across the VF ecosystem, focusing on key areas such as computer vision for disease detection, crop growth and yield prediction, smart climate control, and precision nutrient and irrigation management. This review examines the performance of different algorithms, including Convolutional Neural Networks (CNNs), Random Forest, XGBoost, and LSTMs across hydroponic, aeroponic, and aquaponic systems. The review also covers IoT setups with multi-sensor networks, edge-cloud computing, and automated control systems. Commercial farms have shown real gains in resource efficiency and shorter supply chains. However, challenges remain: high energy use (especially from LED lighting, which makes up 40–60% of costs), expensive setup, scattered datasets, and limited real-world testing. Many high-accuracy claims (>95%) come from lab conditions and need better validation in actual farms. Overall, AI-powered vertical farming has strong potential to support resilient urban food systems. Future work should focus on lightweight edge AI models, improved data standards, explainable AI, and robust life cycle assessments to ensure the benefits outweigh the environmental and economic costs. Full article

Smallholder and family farms in the Western Balkans face intertwined production, market, institutional, human resources and financial risks. This systematic review maps agricultural-risk evidence across seven Western Balkan countries and five enterprise types: crops, livestock, poultry, beekeeping and mixed systems. Following PRISMA 2020, SciSpace was used as a discovery interface for Scopus-indexed literature and was complemented by regional literature compilation, citation tracking and manual verification. From 189 records identified before deduplication, 68 studies published between 2000 and 2025 met the eligibility criteria. Studies were coded by country, enterprise type, risk category and method, and synthesised through a 5 × 5 evidence-gap matrix and complementary risk-overlap analysis. Evidence is concentrated in production risk for crops and livestock, market risk in beekeeping and institutional risk in mixed systems. The main gaps concern human resources risk in crops and poultry and financial risk in poultry. Although 22 studies address multiple risks, only six cover all five major farm risks, mostly descriptively rather than through interaction modelling. The review is limited by uneven regional source visibility and judgement-based evidence coding. The findings support portfolio-based risk management for sustainable rural development rather than isolated hazard responses. The review was not prospectively registered. Full article

Arc faults in low-voltage smart-farm distribution systems are difficult to detect reliably because normal current waveforms vary with load composition and operating conditions. This paper presents a progressive three-stage time–frequency learning framework that identifies series arc faults directly from normalized current waveforms. In Stage I, fast Fourier transform (FFT)-based and wavelet-based learnable front-ends independently learn complementary global spectral and local transient representations. In Stage II, their representation spaces are structured using supervised contrastive learning with view-level and cross-spectrum alignment. In Stage III, the aligned feature extractors are frozen, and only the feature fusion block and final classifier are trained for 10-class classification. Current signals are acquired from a practical smart-farm distribution system and evaluated using a measurement-session-level group-wise split. The framework achieves 99.92% accuracy and 99.93% macro recall while maintaining 1.06 M parameters and a model-only inference latency of 8.70 ms. The results demonstrate that the proposed framework provides robust discrimination across unseen measurement sessions within the evaluated load categories and operating conditions. Full article

Wind power integration into mountainous power grids amplifies probabilistic forecasting challenges arising from strong non-stationarity, multi-source meteorological redundancy and frequent curtailment events. To address the limitations of existing approaches, this paper proposes a multi-horizon probabilistic forecasting framework integrating multi-perspective entropy-weighted fusion, permutation-entropy-guided decomposition, and residual-anchored probability modelling. First, an Entropy-Weighted Multi-criteria Permutation Feature Importance (EW-MPFI) module fuses KSG mutual information, Tree-SHAP, and elastic-net permutation importance through entropy-based weighted aggregation, distilling 23-dimensional meteorological inputs into eight informative features while suppressing single-criterion bias. Then, a three-stage decomposition strategy applies ICEEMDAN primary decomposition, permutation-entropy and sample-entropy guided band reconstruction, and SSA secondary refinement on high-frequency components, achieving complexity-aligned multi-scale separation. Finally, a decomposition-aware patch-based Transformer backbone (DPC-Former) generates three-quantile point forecasts, upon which an NGBoost residual layer models the conditional distribution via natural-gradient optimization in the information-geometric parameter space. Case studies on a 130 MW mountainous wind farm in Sichuan, China, covering 8736 15-min samples with 566 curtailment samples (6.48% of the dataset), show that, under the partition-wise offline batch-evaluation protocol, the proposed framework achieves an NMAE of 5.21%, an NCRPS of 3.74%, and a PICP80 of 0.84 across forecasting horizons from 15 min to 4 h. Ablation analysis attributes NMAE improvements of 25.36% and 24.57% to the decomposition and feature-selection modules, respectively, while 50-seed ensembling further reduces NCRPS, NMAE, and NRMSE by 7.40%, 7.00%, and 13.70% relative to single-seed training. A fixed-checkpoint test-block diagnostic further shows limited sensitivity at approximately weekly and three-day decomposition cadences, but a material degradation at a one-day cadence. The reported metrics should therefore be interpreted as offline best-case results rather than as performance under strictly causal real-time deployment. Full article

This study investigates how offshore wind farm layout geometry influences farm-level performance using a computationally efficient Jensen–Park wake model combined with directionally resolved Weibull wind-speed statistics. A fixed-capacity 1.8 GW case study, consisting of 120 V236-15.0 MW turbines, is used to examine the effects of grid aspect ratio, inter-turbine spacing, cumulative row skew, and global layout rotation on wake losses, annual energy production (AEP), and capacity factor under representative offshore screening assumptions. Structured layouts with identical turbine count and installed capacity are compared with a regular baseline grid to isolate geometric effects within a consistent modelling framework. For the nominal offshore Jensen wake-expansion coefficient, k = 0.04, the highest sampled AEP is obtained for the 5 × 24 configuration, which produces 8930.69 GWh yr⁻¹ and a capacity factor of 56.64%. The regular baseline produces 7397.50 GWh yr⁻¹ and a capacity factor of 46.91%, corresponding to a 20.73% AEP increase for the highest sampled layout. However, the performance differences among Layouts D–F are small, indicating a high-performing layout plateau rather than a clearly separated optimum. The contribution of this paper is therefore not a new wake model, optimisation algorithm, or general offshore design rule. Instead, this study provides an auditable screening workflow that documents modelling assumptions, parameter bounds, coordinate transformations, convergence checks, sensitivity analyses, and spatial-efficiency indicators for one turbine model, one turbine count, one synthetic wind rose, and a limited set of structured row–column layouts. Full article

Accurate identification of eucalyptus plantation maturity, which is hierarchically defined as young forests, middle-aged forests, and mature forests, corresponding to different growth and physiological development stages of eucalyptus stands, is critical for forestry management and sustainable development. Currently, research on fine-grained semantic segmentation of different eucalyptus growth stages remains insufficient, and there is a lack of systematic evaluation of classical deep learning architectures and multimodal data for eucalyptus plantation maturity identification. This limits the application of remote sensing technology in forestry management, and constrains the understanding of growth dynamics in subtropical planted forests. Taking Gaofeng Forest Farm in Nanning, Guangxi, as the study area, this study analyzes the adaptability of four generations of deep learning segmentation architectures—U-Net (convolutional baseline), Trans-UNet (CNN-Transformer hybrid), Swin-UNet (pure Transformer), and Mamba-UNet (state-space model)—in the fine identification of eucalyptus plantation maturity based on spectral indices (SIs), C-band SAR data (S1), and multispectral data (S2). The results show the following: (1) There is no positive correlation between model complexity and recognition performance. Among all architectures, Mamba-UNet achieves the best performance, with a validation set mIoU of 79.10% and an F1-score of 88.26%. The performance ranking of the different architectures evaluated is Mamba-UNet > Swin-UNet > U-Net > Trans-UNet. (2) The S2+SI combination achieves the highest accuracy (mIoU 79.10%), outperforming S2+S1+SI (78.55%), S2+S1 (78.38%), and single S2 (77.72%), which indicates the strong correlation between spectral indices and eucalyptus physiological characteristics. The backscattering features of S1 are limited by canopy penetration in the subtropical rainforest, introducing redundancy and triggering negative fusion effects. (3) Independent verification with field survey points verifies the strong generalization ability of the proposed approach, with OA of 94.08%, mIoU of 85.62%, Precision of 93.00%, Recall of 91.37% and F1-score of 92.15%, which indicates that the methodological framework can realize the identification of eucalyptus maturity with high precision. (4) Eucalyptus accounts for 60.15% of the total area of Gaofeng Forest Farm. Within the eucalyptus stand age structure, young, middle-aged, and mature forests account for 20.67%, 13.62%, and 25.86%, respectively. The overall distribution exhibits a polarized pattern with high proportions of young and mature forests. The findings offer theoretical support and insights for dynamic monitoring of fast-growing plantations and refined management of stand development stages. Full article

Accurate and efficient recognition of tomato leaf diseases and pests is essential for precision agriculture, yet practical deployment remains challenging due to complex backgrounds, domain shifts, and subtle inter-class visual differences. From a broader mathematical perspective, this task can be viewed as token-level representation refinement under noise, ambiguity, and distribution shift. Recent state-space-model-based vision backbones provide favorable efficiency for high-resolution imagery but lack explicit mechanisms for adaptive feature refinement under noisy conditions. To address these issues, we propose EATMamba, an evolution-inspired Vision Mamba framework for tomato leaf disease and pest classification. Rather than being limited to a task-specific classifier, EATMamba is formulated as an evolution-inspired differentiable token-refinement mechanism designed to be compatible with state-space visual recognition backbones. EATMamba introduces two lightweight and fully differentiable modules—Evolutionary Crossover–Interaction and Knowledge-Guided Mutation–Selection—which perform token-level recombination and selective refinement to emphasize discriminative disease cues while suppressing irrelevant background information. These modules are inspired by crossover, mutation, and selection concepts, but are implemented as trainable differentiable operations over visual tokens, forming a generate–recombine–select style mechanism for representation refinement. The scope of this study is low-cost RGB-based visible-symptom disease and pest classification, rather than pre-symptomatic early disease detection. Extensive experiments on two complementary tomato datasets, including a controlled high-resolution dataset and an in-the-wild farm dataset, demonstrate that EATMamba consistently outperforms representative CNN-, Transformer-, and state-space-model-based baselines. Ablation studies and visualization analyses further confirm the complementary contributions of the proposed modules. Overall, EATMamba provides an effective and efficient framework for fine-grained plant disease recognition and illustrates how evolution-inspired principles can be incorporated into modern vision backbones for robust agricultural image analysis. Full article

This paper systematically reviews the emerging field of smart urban agriculture, defined as the integration of information and communication technologies (ICTs) into food production practices within and around built-up urban areas. Although urban agriculture and digital agriculture have each generated substantial scholarship, their intersection remains conceptually fragmented and empirically uneven. Following the PRISMA 2020 guidelines, this study reviews 143 English-language articles indexed in Web of Science and Scopus between 2016 and 2026, combining bibliometric mapping with structured thematic coding. The analysis shows that smart urban agriculture has expanded rapidly since 2021, but remains geographically concentrated and disciplinarily dispersed. Current research is organized mainly around IoT and sensor networks, machine learning, soilless cultivation, vertical farming, plant factories, and controlled-environment agriculture. ICT applications are most mature in enclosed, data-rich, and technically controllable systems, where they support monitoring, prediction, automation, and resource optimization. By contrast, community-based, open-space, and governance-oriented forms of urban agriculture remain underexplored. By systematically linking ICT families with different urban agriculture production settings, this review clarifies the field’s emerging knowledge structure and demonstrates that technological development remains uneven across agricultural forms and socio-institutional contexts. Full article

Vertical farming (VF) holds potential for urban food security in African cities, yet adoption of VF systems by farmers remains poorly understood across diverse urban contexts. This study modeled VF system diffusion across three settlement types in Lagos and Nairobi using the ADOPT Tool framework. A total of 60 urban farmers from the 2 cities participated in the focus group discussions at six sites. The study provides an exploratory, model-based contribution to understanding context-specific adoption of vertical farming in selected African urban settings. Predictions revealed substantial spatial heterogeneity. Peak adoption ranged from 2 to 98%, with time-to-peak spanning 10–25 years. Intra-city variation in basic system adoption exceeded inter-city differences (61–97% in Nairobi versus 16–52% in Lagos). CEA systems approached saturation in affluent neighborhoods (Langata 98%, Lekki 95%) while remaining constrained in informal settlements (<25%). Sensitivity analysis identified three population-level binding constraints: Financial Capacity, Management Skill, and Environmental Orientation. Each constraint generated an 18–24 percentage-point impact in resource-constrained settlements but less than 6 percentage points in affluent areas. On the other hand, four innovation-level factors (Complexity, Trialability, Observability, Training Requirements) accelerated the diffusion rate by 1.2–1.4 years across all sites. These findings demonstrate that adoption is determined by context-specific constraint configurations rather than uniform urban conditions. Effective scaling requires settlement-differentiated interventions: in informal settlements, finance and extension; in high-density areas, space-efficient design and market linkages; in affluent zones, regulatory enablement. The study illustrates that technology adoption models developed in wealthy contexts often fail to apply to resource-constrained African cities where local constraints vary dramatically over short distances. The study also finds that adoption can be improved by implementing targeted policies and strengthening supporting infrastructure. These findings provide actionable pathways to increase vertical farming adoption in rapidly growing African megacities. Full article