Search Results (609)

Evaluating performance is a necessary and specific process across all sectors and organizational levels, shaped by context, indicators, and purpose. Considering global sustainability transitions, understanding financial performance entails a deeper perspective on technical accuracy, conceptual clarity, and systemic integration. This study investigates how financial performance is assessed and interpreted in sustainability-focused research, drawing on a bibliometric analysis of 490 articles indexed in the Web of Science from 2007 to 2023. Using SciMAT, we traced thematic evolutions and revealed a fragmented research landscape marked by competing theoretical, methodological, and practical orientations. To address this conceptual dispersion, we applied the Analytic Hierarchy Process (AHP) to evaluate five key alternatives to financial-performance assessment (quantitative measurement, definition-oriented reasoning, theoretical frameworks, experiential comparison, and integration with sustainability and ethics) against three conceptual criteria (philosophical depth, holistic scope, and multidisciplinary relevance). The results highlight a strong preference for holistic and integrative models of financial performance, with quantitative measurement ranking highest in practical terms, followed by experiential and sustainability-driven approaches. These results underscore the need to align financial evaluation more closely with sustainability values, bridging short-term metrics with long-term societal impact. By combining diachronic thematic mapping with structured decision analysis, this study advances a more reflective and forward-looking framework for performance research. It contributes to sustainability research by identifying underexplored epistemological pathways and supporting the development of financial evaluation models that are inclusive, ethically grounded, and aligned with sustainable development goals. Full article

In recent years, Vision Transformers (ViTs) have gained significant traction in the field of hyperspectral image classification due to their advantages in modeling long-range dependency relationships between spectral bands and spatial pixels. However, after stacking multiple Transformer encoders, challenges pertaining to information degradation may emerge during the forward propagation. That is to say, existing Transformer-based methods exhibit certain limitations in retaining and effectively utilizing information throughout their forward transmission. To tackle these challenges, this paper proposes a novel dual-branch spatial–spectral Transformer model that incorporates similarity propagation (DBSSFormer-SP). Specifically, this model first employs a Hybrid Pooling Spatial Channel Attention (HPSCA) module to integrate global information by pooling across different dimensional directions, thereby enhancing its ability to extract salient features. Secondly, we introduce a mechanism for transferring similarity attention that aims to retain and strengthen key semantic features, thus mitigating issues associated with information degradation. Additionally, the Spectral Transformer (SpecFormer) module is employed to capture long-range dependencies among spectral bands. Finally, the extracted spatial and spectral features are fed into a multilayer perceptron (MLP) module for classification. The proposed method is evaluated against several mainstream approaches on four public datasets. Experimental results demonstrate that DBSSFormer-SP exhibits excellent classification performance. Full article

Background: Current cosmological fits typically assume a direct relation between cosmic time (t) and the scale factor ($a\left(t\right)$), yet this ansatz remains largely untested across diverse observations. Objectives: We (i) test whether a single power-law scaling ($a\left(t\right)\propto t^{\alpha }$) can reproduce late- and early-time cosmological data and (ii) explore whether a dynamically evolving ($\alpha \left(t\right)$), modeled as a scalar–tensor field, naturally induces directional asymmetry in cosmic evolution. Methods: We fit a constant-$\alpha$ model to four independent datasets: 1701 Pantheon+SH0ES supernovae, 162 gamma-ray bursts, 32 cosmic chronometers, and the Planck 2018 TT spectrum (2507 points). The CMB angular spectrum is mapped onto a logarithmic distance-like scale ($\mu ={log}_{10}{D}_{\ell }$), allowing for unified likelihood analysis. Each dataset yields slightly different preferred values for $H_0$ and $\alpha$; therefore, we also perform a global combined fit. For scalar–tensor dynamics, we integrate $\alpha \left(t\right)$ under three potentials—quadratic, cosine, and parity breaking (${\alpha }^{3}sin\alpha$)—and quantify directionality via forward/backward evolution and Lyapunov exponents. Results: (1) The constant-$\alpha$ model achieves good fits across all datasets. In combined analysis, it yields $H_0\simeq 70\mathrm{km}{\mathrm{s}}^{-1}{\mathrm{Mpc}}^{-1}$ and $\alpha \simeq 1.06$, outperforming $\Lambda$CDM globally ($\Delta \mathrm{AIC}\simeq 401254$), though $\Lambda$CDM remains favored for some low-redshift chronometer data. High-redshift GRB and CMB data drive the improved fit. Numerical likelihood evaluations are approximately three times faster than for $\Lambda$CDM. (2) Dynamical $\alpha \left(t\right)$ models exhibit time-directional behavior: under asymmetric potentials, forward evolution displays finite Lyapunov exponents (${\lambda }_L\sim {10}^{-3}$), while backward trajectories remain confined (${\lambda }_L<0$), realizing classical arrow-of-time emergence without entropy or quantum input. Limitations: This study addresses only homogeneous background evolution; perturbations and physical derivations of potentials remain open questions. Conclusions: The time-scaling approach offers a computationally efficient control scenario in cosmological model testing. Scalar–tensor extensions naturally introduce classical time asymmetry that is numerically accessible and observationally testable within current datasets. Code and full data are available. Full article

Photovoltaic (PV) systems are key renewable energy sources due to their ease of implementation, scalability, and global solar availability. Enhancing their lifespan and performance is vital for wider adoption. Identifying degradation root causes is essential for improving PV design and maintenance, thus extending lifespan. This paper proposes a hybrid fault diagnosis method combining a bond graph-based PV cell model with empirical degradation models to simulate faults, and a deep learning approach for root-cause detection. The experimentally validated model simulates degradation effects on measurable variables (voltage, current, ambient, and cell temperatures). The resulting dataset trains an Optimized Feed-Forward Neural Network (OFFNN), achieving 75.43% accuracy in multi-class classification, which effectively identifies degradation processes. Full article

Climate change continues to pose significant challenges to global water security, with dams being particularly vulnerable to hydrological cycle alterations. This study investigated the climate-based impact on the hydrology of the Doğancı dam, located in Bursa, Turkey, using feed-forward neural networks (FNNs). The modeling used meteorological parameters as inputs. The employed FNN comprised one input, hidden, and output layer. The efficacy of the models was evaluated by comparing the correlation coefficients (R), mean squared errors (MSE), and mean absolute percentage errors (MAPE). Furthermore, two training algorithms, namely Levenberg-Marquardt and resilient backpropagation, were employed to determine the algorithm that yields more accurate output predictions. The findings of the study showed that the model using air temperature, solar radiation, solar intensity, evaporation, and evapotranspiration as predictors for the water budget and water level of the Doğancı dam exhibited the lowest MSE (0.59) and MAPE (1.31%) and the highest R (0.99) compared to other models under LM training. The statistical analysis determined no significant difference (p > 0.05) between the Levenberg and Marquardt and resilient backpropagation training algorithms. However, a visual interpretation revealed that the Levenberg-Marquardt algorithm outperformed the resilient backpropagation, yielding lower errors, higher correlation values, and faster convergence for the models tested in this study. The novelty of this study lies in the use of certain meteorological inputs, particularly snow depth, for dam inflow forecasting, which has seldom been explored. Moreover, this study compared two widely used ANN training algorithms and applied the modeling framework to a region of strategic importance for Turkey’s water security. This study highlights the effectiveness of ANN-based modeling for hydrological forecasting and determining climate-induced impacts on water bodies such as dams and reservoirs. Full article

Strapdown gravity systems are increasingly employed in airborne geophysical exploration and geodetic studies due to advantages such as ease of installation, wide dynamic range, and adaptability to various platforms, including airplanes, helicopters, and large drones. This study presents results from an airborne gravity survey conducted over the northeastern sector of Sardinia (Italy), using a high-resolution strapdown gravity ensuring an accuracy of approximately 1 mGal. Data were collected at an average altitude of 1800 m with a spatial resolution of 3.0 km. The survey focused on the Sos Enattos area near Lula (Nuoro province), a candidate site for the Einstein Telescope (ET), a third-generation gravitational wave observatory. The ideal site is required to be geologically and seismically stable with a well-characterized subsurface. To support this, we performed a new gravity survey to complement existing geological and seismic data aimed at characterizing the mid-to-shallow crustal structure of Sos Enattos. Results show that the strapdown system effectively detects gravity anomalies linked to crustal sources down to ~3.5 km, with particular emphasis within the 1–2 km depth range. Airborne gravity data reveal higher frequency anomalies than those resolved by the EGM2008 global gravity model and show good agreement with local terrestrial gravity data. Forward modeling of the gravity field suggests a crust dominated by alternating high-density metamorphic rocks and granitoid intrusions of the Variscan basement. These findings enhance the geophysical understanding of Sos Enattos and support its candidacy for the ET site. Full article

LiDAR point clouds, as direct carriers of 3D spatial information, comprehensively record the geometric features and spatial topological relationships of object surfaces, providing intelligent systems with rich 3D scene representation capability. However, current point cloud semantic segmentation methods primarily extract features through operations such as convolution and pooling, yet fail to adequately consider sparse features that significantly influence the final results of point cloud-based scene perception, resulting in insufficient feature representation capability. To address these problems, a sparse feature dynamic graph convolutional neural network, abbreviated as SFDGNet, is constructed in this paper for LiDAR point clouds of complex scenes. In the context of this paper, sparse features refer to feature representations in which only a small number of activation units or channels exhibit significant responses during the forward pass of the model. First, a sparse feature regularization method was used to motivate the network model to learn the sparsified feature weight matrix. Next, a split edge convolution module, abbreviated as SEConv, was designed to extract the local features of the point cloud from multiple neighborhoods by dividing the input feature channels, and to effectively learn sparse features to avoid feature redundancy. Finally, a multi-neighborhood feature fusion strategy was developed that combines the attention mechanism to fuse the local features of different neighborhoods and obtain global features with fine-grained information. Taking S3DIS and ScanNet v2 datasets, we evaluated the feasibility and effectiveness of SFDGNet by comparing it with six typical semantic segmentation models. Compared with the benchmark model DGCNN, SFDGNet improved overall accuracy $\left(OA\right)$, mean accuracy $\left(mAcc\right)$, mean intersection over union $\left(mIoU\right)$, and $sparsity$ by $1.8\%$, $3.7\%, 3.5\%$, and $85.5\%$ on the S3DIS dataset, respectively. The $mIoU$ on the ScanNet v2 validation set, $mIoU$ on the test set, and $sparsity$ were improved by $3.2\%, 7.0\%$, and $54.5\%$, respectively. Full article

Against the backdrop of the intensifying global climate crisis, urban construction land (UCL), as a major source of carbon emissions, faces the severe challenge of balancing emissions reduction and development in its low-carbon transformation. This study is dedicated to filling the theoretical and methodological gap in the refined assessment of urban construction land carbon effects (UCLCE) spatial heterogeneity among regions, and proposes and validates an innovative block-scale prediction framework. To achieve this goal, this study takes the central urban area of Changxing, Zhejiang Province, as the study area and establishes a BP neural network model for predicting UCLCE based on multi-source data such as building energy consumption and built environment elements (BEF). The results demonstrate that the BP neural network model effectively predicts the different types of UCLCE, with an average error rate of 30.10%. (1) The total effect and intensity effect exhibit different trends in the study area, and a carbon effect table for different types of UCL is established. (2) The spatial distribution characteristics of UCLCE reveal a distinct reverse-L pattern (“┙”-shaped layout) with positive spatial correlation (Moran’s I = 0.11, p < 0.001). (3) The model’s core practical value lies in enabling forward-looking assessment of carbon effects in urban planning schemes and precise quantification of emissions reduction benefits. Optimization trials on representative blocks achieve up to 25.45% carbon reduction. This study provides theoretical foundations for understanding UCLCE spatial heterogeneity while delivering scientifically grounded tools for diagnosing built environment issues and advancing low-carbon optimization in urban renewal contexts. These contributions carry significant theoretical and practical implications. Full article

Underwater vehicles serve as critical assets for global ocean exploration and naval capability enhancement. The marine environment exhibits intricate hydrodynamic phenomena that significantly threaten underwater vehicle navigation safety, particularly in four prevalent complex conditions: surface waves, oceanic currents, stratified fluids, and internal waves. This comprehensive review systematically examines the impacts of these four marine environments on underwater vehicles through critical analysis and synthesis of contemporary advances in theoretical frameworks, experimental methodologies, and numerical simulation approaches. The identified influences are categorized into five primary aspects: hydrodynamic characteristics, dynamic response patterns, load distribution mechanisms, navigation trajectory optimization, and stealth performance. Particular emphasis is placed on internal wave interactions, with rigorous analysis derived from experimental investigations and numerical modeling of internal wave dynamics and their coupling effects with underwater vehicles. In addition, this review points out and analyzes the shortcomings of the current research in various aspects and puts forward some thoughts and suggestions for future research directions that are worth further exploration, including enriching the research objects, upgrading the experimental techniques, and introducing artificial intelligence methods. Full article

A long-wavelength geoidal geometry reflects mainly lateral density variations in the Earth’s mantle, with the most pronounced features of the Indian Ocean Geoid Low and the West Pacific and North Atlantic Geoid Highs. Despite this spatial pattern being clearly manifested in the global geoidal geometry determined from gravity-dedicated satellite missions, the gravitational signature of the deep mantle could be refined by modelling and subsequently removing the gravitational contribution of lithospheric geometry and density structure. Nonetheless, the expected large uncertainties in available lithospheric density models (CRUST1.0, LITHO1.0) limit, to some extent, the possibility of realistically reproducing the gravitational signature of the deep mantle. To address this issue, we inspect an alternative approach. Realizing that the gravity geopotential field (i.e., gravity potential) is smoother than its gradient (i.e., gravity), we apply the integral operator to geopotential and then investigate the spatial pattern of this functional (i.e., radially integrated geopotential). Results show that this mathematical operation enhances a long-wavelength signature of the deep mantle by filtering out the gravitational contribution of the lithosphere. This finding is explained by the fact that in the definition of this functional, spherical harmonics of geopotential are scaled by the factor 1/n (where n is the degree of spherical harmonics), thus lessening the contribution of higher-degree spherical harmonics in the radially integrated geopotential. We also demonstrate that further enhancement of the mantle signature in this functional could be achieved based on modelling and subsequent removal of the gravitational contribution of lithospheric geometry and density structure. Full article

In this paper, we develop an HIV/AIDS epidemic model that incorporates a saturated incidence rate to reflect the limited transmission capacity and the impact of behavioral saturation in contact patterns. The model is formulated as a system of seven non-linear ordinary differential equations representing key population compartments. In addition to model formulation, we introduce an optimal control problem involving three control measures: educational campaigns, screening of unaware infected individuals, and antiretroviral treatment for aware infected individuals. We begin by establishing the positivity and boundedness of the model solutions under constant control inputs. The existence and local and global stability of both the disease-free and endemic equilibrium points are analyzed, depending on the effective reproduction number (${\mathcal{R}}_e$). Bifurcation analysis reveals that the model undergoes a forward bifurcation at ${\mathcal{R}}_e=1$. A local sensitivity analysis of ${\mathcal{R}}_e$ identifies the disease transmission rate as the most sensitive parameter. The optimal control problem is then formulated by incorporating the dynamics of infected subpopulations, control costs, and time-dependent controls. The existence of optimal control solutions is proven, and the necessary conditions for optimality are derived using Pontryagin’s Maximum Principle. Numerical simulations support the theoretical analysis and confirm the stability of the equilibrium points. The optimal control strategies, evaluated using the Incremental Cost-Effectiveness Ratio (ICER), indicate that implementing both screening and treatment (Strategy D) is the most cost-effective intervention. These results provide important insights for designing effective and economically sustainable HIV/AIDS intervention policies. Full article

Deep learning has demonstrated significant success in hyperspectral image (HSI) classification by effectively leveraging spatial–spectral feature learning. However, current approaches encounter three challenges: (1) high spectral redundancy and the presence of noisy bands, which impair the extraction of discriminative features; (2) limited spatial receptive fields inherent in convolutional operations; and (3) unidirectional context modeling that inadequately captures bidirectional dependencies in non-causal HSI data. To address these challenges, this paper proposes HG-Mamba, a novel hybrid geometry-aware bidirectional Mamba network for HSI classification. The proposed HG-Mamba synergistically integrates convolutional operations, geometry-aware filtering, and bidirectional state-space models (SSMs) to achieve robust spectral–spatial representation learning. The proposed framework comprises two stages. The first stage, termed spectral compression and discrimination enhancement, employs multi-scale spectral convolutions alongside a spectral bidirectional Mamba (SeBM) module to suppress redundant bands while modeling long-range spectral dependencies. The second stage, designated spatial structure perception and context modeling, incorporates a Gaussian Distance Decay (GDD) mechanism to adaptively reweight spatial neighbors based on geometric distances, coupled with a spatial bidirectional Mamba (SaBM) module for comprehensive global context modeling. The GDD mechanism facilitates boundary-aware feature extraction by prioritizing spatially proximate pixels, while the bidirectional SSMs mitigate unidirectional bias through parallel forward–backward state transitions. Extensiveexperiments on the Indian Pines, Houston2013, and WHU-Hi-LongKou datasets demonstrate the superior performance of HG-Mamba, achieving overall accuracies of 94.91%, 98.41%, and 98.67%, respectively. Full article

With the increasing frequency of extreme weather events, global concerns regarding climate change have intensified, with carbon dioxide widely recognized as the primary driver of global warming and climate disruption. It is necessary to investigate how to develop industries to meet the constant GDP growth and minimum carbon emissions. This study investigates the optimization of industrial structure under China’s ‘Dual Carbon’ Goal in Guangdong Province from 2012 to 2017, employing a multi-objective programming model. Using the input–output table, carbon emissions across 42 industries are calculated based on the Intergovernmental Panel on Climate Change (IPCC) carbon emission factor method. According to Hirschman’s theory of industrial interdependence, the economic and carbon emission linkage coefficients between these industries are obtained by calculating the Ghosh inverse matrix and the Leontief inverse matrix to analyze the economic forward and backward linkage of the industries, as well as the carbon emission forward and backward linkage. The impact of industry input and output on the urban economy and the resulting carbon emission problems are discussed, and industries are divided into encouraged and restricted industries. Using a multi-objective programming model, the expected final demand, changes in final demand, and expected carbon emissions of these industries under the ‘Dual Carbon’ Goal, with the target of maintaining the same economic growth rate and promoting carbon reduction, are analyzed. The results show that most industries in Guangdong Province need to reduce final demand, including the highest carbon-emitting industries and industries that are relatively restricted by scale in development. The policy implications of optimizing the industrial structure to reduce carbon emissions are provided. Full article

The global burden of diabetes mellitus (DM) continues to escalate, posing significant challenges to healthcare systems worldwide. This study compares machine learning (ML) and deep learning (DL) methods, their hybrids, and ensemble strategies for predicting the health outcomes of diabetic patients. This work aims to find the best solutions that strike a balance between computational efficiency and accurate prediction. The study systematically assessed a range of predictive models, including sophisticated DL techniques and conventional ML algorithms, based on computational efficiency and performance indicators. The study assessed prediction accuracy, processing speed, scalability, resource consumption, and interpretability using publicly accessible diabetes datasets. It methodically evaluates the selected models using key performance indicators (KPIs), training times, and memory usage. AdaBoost had the highest F1-score (0.74) on PIMA-768, while RF excelled on PIMA-2000 (~0.73). An RNN led the 3-class BRFSS survey (0.44), and a feed-forward DNN excelled on the binary BRFSS subset (0.45), while RF also achieved perfect accuracy on the EMR dataset (1.00) confirming that model performance is tightly coupled to each dataset’s scale, feature mix and label structure. The results highlight how lightweight, interpretable ML and DL models work in resource-constrained environments and for real-time health analytics. The study also compares its results with existing prediction models, confirming the benefits of selected ML approaches in enhancing diabetes-related medical outcomes that are substantial for practical implementation, providing a reliable and efficient framework for automated diabetes prediction to support initiative-taking disease management techniques and tailored treatment. The study concludes the essentiality of conducting a thorough assessment and validation of the model using current institutional datasets as this enhances accuracy, security, and confidence in AI-assisted healthcare decision-making. Full article

Achieving global net-zero emissions by 2050 demands integrated and scalable strategies that unite decarbonization technologies across sectors. This review provides a forward-looking synthesis of carbon capture and storage and hydrogen systems, emphasizing their integration through artificial intelligence to enhance operational efficiency, reduce system costs, and accelerate large-scale deployment. While CCS can mitigate up to 95% of industrial CO₂ emissions, and hydrogen, particularly blue hydrogen, offers a versatile low-carbon energy carrier, their co-deployment unlocks synergies in infrastructure, storage, and operational management. Artificial intelligence plays a transformative role in this integration, enabling predictive modeling, anomaly detection, and intelligent control across capture, transport, and storage networks. Drawing on global case studies (e.g., Petra Nova, Northern Lights, Fukushima FH2R, and H21 North of England) and emerging policy frameworks, this study identifies key benefits, technical and regulatory challenges, and innovation trends. A novel contribution of this review lies in its AI-focused roadmap for integrating CCS and hydrogen systems, supported by a detailed analysis of implementation barriers and policy-enabling strategies. By reimagining energy systems through digital optimization and infrastructure synergy, this review outlines a resilient blueprint for the transition to a sustainable, low-carbon future. Full article