Search Results (804)

Rolling element bearings are critical components in rotating machinery, yet assessing and predicting their reliability under heavy industrial noise remains challenging. Existing methods suffer from three major limitations: (1) single-channel signal processing and single-scale indicators lack robustness against non-stationary noise; (2) classical multi-channel decomposition methods, such as multivariate variational mode decomposition (MVMD), rely on empirical parameter tuning, which frequently leads to over- or under-decomposition; and (3) monolithic deep architectures and homogeneous ensemble models suffer from prediction drift and generalization bottlenecks during long-term temporal extrapolation. To address these issues, this paper introduces an automated framework that combines adaptive multi-channel signal purification with a heterogeneous stacking ensemble (HeteroStack-LR). Unlike conventional MVMD pipelines that fix $[K,\alpha ]$ empirically, the Sequoia Optimization Algorithm (SOA) autonomously determines the globally optimal configuration, achieving a mean SNR of $-2.08\mathrm{dB}$—a $1.88$ to $2.50\mathrm{dB}$ improvement over standard VMD/MVMD baselines—along with up to a $37.1\%$ reduction in computation time. Rather than relying on conventional single-metric intrinsic mode function (IMF) selection, we construct a multi-domain hybrid index integrating the Fault Correlation Factor, energy ratio, and refined composite multiscale dispersion entropy (RCMDE) to robustly identify noise-resistant components, thereby enhancing denoising quality by $22.4\%$ to $32.2\%$ over single-scale criteria. Furthermore, contrasting with linear PCA-based reduction, Diffusive Topology Neighbor Embedding (D-TNE) effectively preserves the nonlinear manifold structure of degradation trajectories in a low-dimensional space. Finally, a heterogeneous stacked ensemble featuring an out-of-fold (OOF) leakage-prevention strategy and a logistic regression meta-learner is designed to suppress prediction drift while avoiding the over-parameterization typical of deep architectures. Experimental results across four bearing datasets demonstrate that HeteroStack-LR achieves a minimal MAE of $0.063$ with a variance of ≤±0.002, outperforming state-of-the-art deep architectures (such as TCN, CNN-LSTM, BiLSTM-Attention, and Transformer) as well as classical baselines (Bi-LSTM, CNN, and LSSVM). Ablation studies confirm that removing SOA and MVMD degrades MAE by $12.7\%$ and $19.0\%$, respectively, validating that the framework’s strength stems not from any isolated module, but from the end-to-end synergistic integration of signal purification and reliability prediction. Full article

Villages within Key Ecological Function Zones (KEFZs) are governed by a common ecological-priority regime, yet their rural sustainable development potential (RSDP) continues to diverge. This suggests that ecological conditions are not sufficient determinants of rural development outcomes, but prerequisite boundaries within which social livability and economic sustainability are activated. Existing SDG localization studies often treat ecological, social, and economic dimensions as parallel and compensable components, making it difficult to identify the structural sources of village-level differentiation in ecologically constrained regions. Using 1108 administrative villages in the Leishui River Basin as analytical units, this study constructs a village-scale RSDP assessment framework based on panel data for 2010, 2015, 2020, and 2024. Ecological health is positioned as the prerequisite constraint layer, while social livability and economic sustainability are treated as development capacities. A three-dimensional coupling coordination degree model is used to measure the structural matching among ecological foundation, social support, and economic transformation capacity. Spatial autocorrelation, exploratory space–time data analysis, and K-means clustering are further applied to identify spatial dependence, evolutionary trajectories, and bottleneck types. The results show that the mean coordination level of RSDP increased from 0.5807 to 0.6522 during 2010–2024, but this improvement was driven by uneven subsystem trajectories. Economic sustainability improved most rapidly, and ecological health increased steadily, whereas social livability remained the main shortfall. Spatially, RSDP showed significant positive autocorrelation, but clustering intensity weakened and both high- and low-value clusters became more fragmented. The dominant constraint shifted from economic weakness to social-support deficits, indicating that population support, public services, and everyday living capacity became key bottlenecks once basic economic conditions improved. Four village types were identified: suburban integration-led, vulnerable fluctuation-constrained, stable balanced-development, and ecological-resource potential. By reframing ecological conditions as prerequisite constraints rather than parallel assessment dimensions, this study extends SDG-based rural sustainability assessment from composite level comparison to structural potential diagnosis and provides village-scale evidence for differentiated governance in KEFZs. Full article

Medical image segmentation is a critical task in computer-assisted diagnosis, but accurate delineation remains challenging in sensor-captured dermoscopic and endoscopic images because of low contrast, blurred boundaries, acquisition artifacts, and large appearance variations. Although CNN-based methods are effective in capturing local texture and boundary cues, they often struggle to explicitly model long-range dependencies and global structural relationships. Transformer-based architectures can capture global context, but their self-attention mechanism may become computationally costly when processing high-resolution feature maps. To address these challenges, we propose MGA-UNet, a frequency-aware multi-scale encoder–decoder segmentation framework that integrates wavelet-based frequency decomposition with Mamba-based long-range dependency modelling. Specifically, the Wavelet-Mamba feature extraction backbone (WMB) decomposes features into low- and high-frequency components to enhance boundary-aware representation, the Gated Multi-scale Aggregation Module (GMAM) aggregates parallel multi-scale encoder features and applies a content-dependent gate to the fused response, and the Adaptive Sparse Attention Module (ASAM) refines bottleneck representations with sparse attention for global semantic modelling. Across three independent runs with random seeds 42, 123, and 2026, MGA-UNet achieves mean Dice Similarity Coefficients of $88.92\pm 0.04\%$, $88.01\pm 0.07\%$, and $85.91\pm 0.04\%$ on ISIC2018, ISIC2017, and Kvasir-SEG, respectively. These results demonstrate competitive segmentation performance among the compared representative CNN-based, Transformer-based, and Mamba-based methods, including the recent H-VMUNet baseline. These results indicate that frequency-domain decomposition and state-space modelling can complement each other for accurate medical image segmentation, particularly in images with ambiguous boundaries and complex background interference. Full article

Digital transformation has emerged as one of the most influential drivers of contemporary socio-economic change, shaping development trajectories, social resilience, and the capacity of societies to adapt to external shocks. Despite the growing body of research on digital transformation, relatively little attention has been devoted to societal digitalization as a distinct analytical category. Existing approaches remain largely focused on technological infrastructure, economic performance, or organizational transformation, while the social dimension of digital development is frequently treated as secondary. Addressing this gap, the present study conceptualizes societal digitalization as an autonomous dimension of digital transformation and advances a human-centred perspective that emphasizes digital capabilities and meaningful technology use rather than mere access to technological resources. The study examines the digital development trajectories of Indonesia, Malaysia, and Thailand (ASEAN-3) between 2016 and 2023, with particular attention to the transformative effects of the COVID-19 pandemic. To this end, an original Digital Development of Society (DDS) Index was developed and applied. The index is grounded in a hierarchical framework encompassing three interrelated dimensions: Access, Skills, and Use. The findings reveal substantial cross-country differences in both the level and structure of societal digitalization. More importantly, they provide empirical evidence of a second-level digital divide, demonstrating that improvements in digital access do not automatically translate into higher levels of digital competence or more advanced forms of technology utilization. The results further indicate that a structural shift in digital development—moving the focus from connectivity towards digital skills and meaningful use—accelerated during the 2020–2023 period. Consequently, human capital emerges as a more decisive determinant of digital maturity than infrastructure alone. The study contributes to the literature by offering a new conceptual framework for understanding societal digitalization and by introducing a multidimensional measurement tool capable of identifying structural bottlenecks in socio-digital development. Furthermore, the findings extend the policy debate on digital transformation by providing a diagnostic framework that enables policymakers to identify structural bottlenecks in national digital ecosystems and to align infrastructure investments with human capital development and meaningful digital participation. Full article

Industrial robots, particularly six-axis serial manipulators, have been widely deployed in manufacturing workflows including assembly, welding, material handling, inspection and precision machining. As the demand for higher end-effector positioning accuracy and trajectory tracking performance grows, end-position errors induced during manipulator operation—stemming from geometric deviations, joint friction, load fluctuations, current surges, as well as variations in velocity and acceleration—have emerged as a critical bottleneck limiting high-precision applications. Conventional error compensation approaches mostly rely on geometric calibration, empirical formulas or fixed regression algorithms, which struggle to adequately characterize error trends featuring strong temporal dependencies, nonlinearity and multi-factor coupling. To address the aforementioned limitations, this paper takes the UR5 industrial manipulator as the research object. Leveraging the NIST-released dataset for manipulator positional accuracy degradation monitoring, this study develops and implements a physics-aware Transformer-based compensation framework that integrates a physics-consistent constraint loss and a nonlinear exponential error amplification strategy with a standard Transformer encoder for end-effector positional accuracy degradation. Multiple variables including target joint position, velocity, acceleration, torque, motor current and control current are selected to construct time-window input vectors, which are used to train the Transformer regression model to capture the correlation between historical motion states and real-time end-effector positional accuracy degradation. Experimental results demonstrate that the proposed Transformer model can fully capture temporal contextual correlations and multi-feature fusion information embedded within manipulator kinematic data, delivering superior error compensation performance for the six-dimensional end-effector pose error prediction task. The self-attention-based time-series modeling framework is well-suited to the nonlinear, coupled and time-varying characteristics of manipulator operational errors. This work provides valuable references for accuracy enhancement of industrial robots and the design of intelligent error compensation schemes. This work provides valuable references for accuracy enhancement of industrial robots and the design of intelligent error compensation schemes, with the proposed physics-aware strategies being model-agnostic and potentially extensible to other regression architectures. Full article

The trade-off between flowability and anti-dispersion properties of alkali-activated slag slurry in underwater environments represents a key technical bottleneck limiting their application in marine underwater engineering. In this study, granulated blast furnace slag (GGBS) was used as the raw material, with the modulus of liquid sodium silicate adjusted by NaOH serving as the alkali activator, and hydroxypropyl methylcellulose (HPMC) and polyacrylamide (PAM) selected as anti-dispersion agents. This study systematically investigates the synergistic regulation mechanisms of the anti-dispersion agents’ type, dosage, and activator on the anti-dispersion properties and rheological behavior of alkali-activated slag slurry, and revealed the evolution mechanisms of the microstructure of the hardened slurry through X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR), scanning electron microscopy (SEM) and mercury intrusion porosimetry (MIP) analysis. The results indicate that the activator is the key factor in regulating the various properties of the slurry, and the combination of PAM and HPMC produces a significant synergistic effect. At the optimal formulation (modulus of 1.0, total blended anti-dispersion agent content of 1%, and a mass ratio of PAM to HPMC of 1:1), the slurry exhibited a wet loss rate of 38.45%, a solid retention rate of 89.98%, a flow value of 195 mm, and a 28-day compressive strength of 41.62 MPa, achieving an optimal balance between anti-dispersion performance and workability. Full article

Plants have coevolved with microbes for nearly 500 million years; however, their interrelationship is not well understood. Plant roots have an intricate relationship with soil microbes. Such relationships mold the growth, development, immunity and physiology of plants and thus are of immense interest to agriculture and the environment. Technological advancements in sequencing, imaging, omics, synthetic biology and artificial intelligence have allowed scientists to dissect such relationships to a higher resolution. Thus, these advances have facilitated a deeper understanding of plant–microbe interactions and their role in the life cycle of plants and the environment. However, factors such as microbial diversity, microbial abundance, heterogeneity of soil and plasticity of the environment have precluded a clear in situ understanding of microbial community structure. Thus, constructing synthetic microbial communities or SynComs and investigating their effect on plants in a controlled environment offers a reductionist and manageable approach to understand plant–microbe relationships. This approach reduces the confounding variables present in the natural environment and facilitates the understanding of such complex interactions. Members of such communities are identified using 16S rRNA sequencing and are constructed using few microorganisms; often fungal strains are added for cross-kingdom SynComs. Metabolic modeling, metabolic cross-feeding along with ecological and evolutionary principles, can be used while choosing candidates for SynComs. Scalability, transferability, reproducibility, predictability and stability are the major bottlenecks in SynCom research. This emerging area of science may have transformative impact in agriculture, environment and space colonization. In this article, we present our perspective on the latest advancements, challenges and future potential of this technology. Full article

Emerging industrial agglomeration drives green transformation through knowledge spillovers and economies of scale. However, its effects exhibit pronounced nonlinearity and heterogeneity, shaped by spatial externalities and development stages. This paper investigates 19 Chinese urban agglomerations over the period 2014 to 2023. Kernel density estimation based on enterprise-level Point-of-Interest (POI) data is used to characterize spatial agglomeration patterns across eight emerging sectors. A two-stage dynamic network super-efficiency SBM model decomposes Green Transition Efficiency (GTE) into resource utilization and pollution control sub-stages. An XGBoost-SHAP-GEO analytical framework, combined with partial dependence analysis, then identifies nonlinear driving mechanisms. The main findings are as follows: First, emerging industrial agglomeration intensifies and polarizes toward the eastern coast, whereas GTE displays a “high-west, low-east” pattern. This produces a significant spatial mismatch, rooted in the near-saturation of environmental carrying capacity in eastern regions, where congestion effects exceed knowledge spillover dividends. Second, geographic characteristics constitute the primary factor shaping GTE and operate through nonlinear interactions with industrial agglomeration and R&D investment. Notably, their moderation direction is reversible, suggesting that geographic endowments should be understood as “conditional assets” rather than fixed advantages. Third, nonlinear patterns across sectors are highly heterogeneous. The bio-industry is the only sector to achieve a J-shaped positive breakthrough. Information technology and new materials exhibit persistent inhibition, while related services display an extremely narrow threshold window with the deepest negative reversal. Thus, “moderate agglomeration” is a multidimensional concept that shifts dynamically with industry type and regional endowment. Fourth, driving mechanisms display stage-dependent evolution. The incubation stage relies on natural endowments and basic industrial pull, with the green bottleneck residing in resource utilization efficiency. The growth stage faces multiple tensions from coexisting positive and negative effects. The optimization stage shifts toward R&D innovation and industrial greening, marking a qualitative transformation from MAR externalities to Jacobs externalities. In addition, the non-significant linear coefficient in the 2SLS instrumental variable test is consistent with the inverted U-shaped nonlinear finding, further validating the necessity of a nonlinear analytical framework. These findings provide differentiated governance evidence for balancing industrial agglomeration with green sustainable development. Full article

In the context of the global green development wave and the rapid iteration of digital technology, digital empowerment has become the core driving force for high-end manufacturing enterprises to achieve green transformation. At present, China’s manufacturing industry is facing the dual pressures of tightening resource and environmental constraints and industrial upgrading. How to break the bottleneck of green development through digital technology innovation has become a key issue to be solved urgently. Based on the techno-economic paradigm, green development theory and value creation theory, this study constructs a theoretical analysis framework for the green development of a digital-enabling manufacturing industry and deeply analyzes the mechanisms of digital technology (such as big data, Internet of Things, artificial intelligence, etc.) in optimizing energy allocation, improving production efficiency and reducing environmental emissions. By selecting 303 manufacturing enterprises of different scales in China as samples, the structural equation model is used for empirical tests. The results show that (1) digital empowerment has a significant positive impact on the green value performance of manufacturing enterprises, and (2) green development plays an intermediary role between digital empowerment and the green value performance of enterprises; that is, digital technology indirectly promotes green development by improving energy conservation and emission reduction, green innovation and green upgrading of enterprises. The research reveals the internal logic of digitally enabling the green development of Chinese manufacturing enterprises and provides a theoretical basis and implementation path for enterprises to formulate the innovation mechanism of digital–green development. Full article

Promoting digital transformation is one of the important directions for the logistics industry to break through the bottleneck of development and achieve high-quality development. This study takes the logistics enterprises listed in China’s A-share market from 2017 to 2024 as the research sample and uses the two-way fixed-effect model and intermediary-effect model to systematically investigate the impact of digital transformation on the operational efficiency of logistics enterprises and its mechanism. The research draws the following conclusions. First, digital transformation is positively associated with the operational efficiency of logistics enterprises and can pass a variety of robustness tests. Second, the mechanism test shows that innovation capability plays an important intermediary role between digital transformation and the operational efficiency of logistics enterprises. Third, the heterogeneity analysis found that for small-scale enterprises and enterprises in the eastern region, the positive impact is more prominent. Therefore, the actual effect of digital transformation is highly dependent on the enterprise’s own attributes and external environment. Policy making needs to be tailored to local conditions and accurately classified. Full article

Detecting microcracks in aero-engine blades is paramount for ensuring flight safety, yet conventional inspection methods exhibit significant limitations in precision, efficiency, and applicability. This paper proposes a multi-modal fusion detection network for blade microcracks—VUFNet (visible and ultraviolet multi-modal fusion detection network)—which accurately identifies blade microcracks by integrating visible and ultraviolet image features. First, the blade surface undergoes chromic acid anodisation to enhance microcrack features. Subsequently, visible and ultraviolet light illumination is applied to construct a microcrack dataset. VUFNet employs dual backbone networks to extract visible and ultraviolet features. It incorporates CSPSTR (cross-stage partial bottleneck with swin transformer) to optimise feature extraction capability, and VUFM (visible–ultraviolet multi-modal adaptive fusion module) to deeply fuse visible and ultraviolet features. Finally, MFConv (multi-branch fusion convolution) enhances detection accuracy in microcrack target regions. On the held-out test subset acquired from a single compressor-blade type under a fixed laboratory imaging protocol, VUFNet achieved mAP_0.5 and mAP_0.5:0.95 values of 96.8% and 85.5%, respectively, with a detection speed of 52.7 fps. These results demonstrate improved within-domain detection performance over the evaluated baselines. Ablation studies further validate the synergistic optimisation effect of VUFM, CSPSTR, and MFConv on model performance. Full article

Lotus corniculatus is a superior leguminous forage with multiple values including forage, ecological, ornamental and medicinal uses. It is also an ideal material for plant bioreactors. As a core technical approach, genetic transformation overcomes the constraints of traditional breeding and facilitates the targeted improvement in stress resistance and agronomic traits in this species. This review summarizes the research progress of the Agrobacterium-mediated genetic transformation of L. corniculatus, focusing on key procedures such as explant selection, strain selection, infection and co-cultivation regimes, basal medium composition, phytohormone regulation, as well as bacteria elimination and transformant screening strategies. We further elaborate on the applications of this transformation system in enhancing tolerance to abiotic stresses (salt, drought and heat), regulating quality-related traits, and developing plant-based vaccine bioreactors. Additionally, this paper critically discusses the major bottlenecks and challenges constraining existing genetic transformation systems in L. corniculatus, and evaluates the prospects for establishing high-efficiency and genetically stable transformation platforms. This review aims to provide theoretical foundations and technical references for germplasm innovation, molecular breeding and comprehensive utilization of L. corniculatus. Full article

Accurate wind speed forecasting plays a crucial role in the safe and stable operation of wind farms and the efficient integration of renewable energy into modern power systems. However, wind speed exhibits complex spatiotemporal variations affected by diverse meteorological conditions, making high-precision prediction a long-standing bottleneck in wind power scheduling. This paper develops a U-shaped spatiotemporal feature fusion network named U-STNet, which realizes joint modeling of inter-turbine spatial correlations and multi-period long-range temporal dependencies. The model maps raw wind speed series to high-dimensional embeddings and adopts an encoder–decoder U-shaped architecture to complete feature encoding, reconstruction and multi-scale feature extraction, which effectively captures the inherent periodic and seasonal patterns of wind speed. Experiments on the SDWPF wind farm dataset show that U-STNet obtains competitive prediction accuracy across all multi-step forecasting horizons. Compared with traditional statistical models, recurrent neural networks and state-of-the-art Transformer baselines, the proposed method exhibits more stable error accumulation characteristics and superior long-step prediction performance. This verifies the effectiveness of jointly modeling turbine spatial topology and multi-scale temporal dependencies for wind speed forecasting. Full article

Mycotoxin contamination remains a persistent threat to food and feed safety owing to the chemical stability of many mycotoxins, frequent co-occurrence, and matrix-dependent risks. Enzymatic detoxification enables structure-targeted transformation of toxicity-determining motifs, such as epoxide rings, reactive double bonds, amide linkages, and lactone structures. However, free mycotoxin-degrading enzymes are often constrained by poor operational stability, difficult recovery, and limited adaptability to complex matrices. Metal–organic frameworks (MOFs) provide programmable microenvironments for enzyme immobilization through tunable pore structures, interfacial chemistry, and confinement effects. This review links toxic structural motifs with enzymatic transformation targets, discusses MOF–enzyme interface engineering and representative host–enzyme compatibility, and evaluates application modes including single-enzyme systems, multi-enzyme co-immobilization or cascade systems, adsorption–degradation coupling, and detection–degradation integration. Key bottlenecks involving enzyme leakage, mass-transfer limitation, real-matrix stability, scalable preparation, and biosafety are critically discussed. Rather than treating MOFs as passive enzyme carriers, this review proposes an application-oriented framework that integrates toxin structure, enzyme function, MOF interface regulation, matrix compatibility, and safety validation to guide the development of MOF-immobilized degrading enzymes for practical mycotoxin detoxification. Full article

The global fermentation industry faces persistent bottlenecks in scaling laboratory innovations to industrial production, and the integration of synthetic biology (SynBio) and artificial intelligence (AI) within the Design–Build–Test–Learn (DBTL) loop has yielded inconsistent industrial outcomes. This review proposes that transformative impact requires a “scale-up-gated DBTL” framework, in which explicit decision gates constrain every iteration. At the Design phase, scale-down simulation data must inform genetic design choices. At the Test phase, downstream processing compatibility and industrial robustness metrics are enforced as non-negotiable evaluation criteria. At the Learn phase, techno-economic analysis (TEA) and life-cycle assessment (LCA) serve as the convergence criteria, replacing traditional titer plateaus. Through a qualitative cross-sectoral analysis of food, pharmaceutical, agricultural, and energy fermentation, the analysis reveals that workflows incorporating such constraints consistently bridge the valley of death, whereas unconstrained DBTL systematically converges on laboratory optima that are industrially unviable. Five strategic priorities are outlined—embedding TEA/LCA into DBTL, adopting scale-down simulation, building open fermentation data repositories, harmonizing regulatory frameworks, and fostering cross-disciplinary training—as prerequisites for progressing toward fully autonomous, scale-up-aware biomanufacturing. Full article