Search Results (40,558)

Nonlinear systems, characterized by their complex and often unpredictable dynamics, are essential in various scientific and engineering applications. However, accurately modeling these systems remains challenging due to their nonlinearity, high-dimensional interactions, and the privacy concerns inherent in data-sensitive domains. Existing federated learning approaches struggle to model such complex behaviors, particularly due to their inability to capture high-dimensional interactions and their failure to maintain privacy while ensuring robust model performance. This paper presents a novel federated learning framework for the robust approximation of nonlinear systems, addressing these challenges by integrating differential privacy to protect sensitive data without compromising model utility. The proposed framework enables decentralized training across multiple clients, ensuring privacy through differential privacy mechanisms that mitigate risks of information leakage via gradient updates. Advanced neural network architectures are employed to effectively approximate nonlinear dynamics, with stability and scalability ensured by rigorous theoretical analysis. We compare our approach with both centralized and decentralized federated models, highlighting the advantages of our framework, particularly in terms of privacy preservation. Comprehensive experiments on benchmark datasets, such as the Lorenz system and real-world climate data, demonstrate that our federated model achieves comparable accuracy to centralized approaches while offering strong privacy guarantees. The system efficiently handles data heterogeneity and dynamic nonlinear behavior, scaling well with both the number of clients and model complexity. These findings demonstrate a pathway for the secure and scalable deployment of machine learning models in nonlinear system modeling, effectively balancing accuracy, privacy, and computational performance. Full article

A sustainable food system can be described as a system that ensures economic, social, and environmental sustainability to secure food and nutrition for current and future generations. The present research aimed to identify and characterize sustainable food initiatives to understand the governance framework, the motivations, the reasons for success, the typology of actors involved, and future prospects. Semi-structured interviews were conducted for seven initiatives between June and August 2024. The interviews were performed online, recorded, transcribed, and analyzed. The results showed that the main drivers of the initiatives relate to food, sustainability, and economic and social dimensions. Moreover, one of the reasons for the success of the initiatives is the availability and commitment of the people involved. These initiatives were created to meet local needs and promote sustainability as well as to encourage economic circularity, knowledge sharing, rural and local valorization, and waste management. Full article

Narrow-linewidth lasers play a crucial role in nonlinear optics, atomic physics, optical metrology, and high-speed coherent optical communications. Precise linewidth measurement is essential for assessing laser noise characteristics; however, conventional methods are often bulky, costly, and unsuitable for integrated applications. This paper presents a compact and cost-effective delay self-homodyne system for laser linewidth measurement, leveraging a field-programmable gate array (FPGA)-based data acquisition circuit. By employing fast Fourier transform (FFT) analysis, the system achieves high-precision linewidth measurement in the kHz range. Additionally, by optimizing the fiber length, the system effectively suppresses low-frequency and 1/f noise, providing an integrated and efficient solution for advanced laser characterization with enhanced performance and reduced cost. Full article

Federated learning (FL) has emerged as a promising framework for training shared models across diverse participants, ensuring data remains securely stored on local devices. Despite its potential, FL still faces some critical challenges, including data heterogeneity, privacy risks, and substantial communication overhead. Current privacy-preserving FL research frequently fails to tackle complexities posed by heterogeneous data adequately, hence increasing communication expenses. To tackle these issues, we propose a top-k shuffled differential privacy FL (TopkSDP-FL) framework tailored to heterogeneous data environments. To address the model drift issue effectively, we design a novel regularization for local training, drawing inspiration from contrastive learning. To enhance efficiency, we propose a bidirectional top-k communication mechanism that reduces uplink and downlink overhead while strengthening privacy protection through double amplification with the shuffle model. Additionally, we shuffle all local gradient parameters at the layer level to address privacy budget concerns associated with high-dimensional aggregation and repeated iterations. Finally, a formal privacy analysis confirms the privacy amplification effect of TopkSDP-FL. The experimental results further demonstrate its superiority over other state-of-the-art FL methods, with an average accuracy improvement of 3% compared to FedAvg and other leading algorithms under the non-IID scenario, while also reducing communication costs by over 90%. Full article

The emission of toxic gases such as NO₂, NO, SO₂, and CO from industrial activities, transportation, and energy production poses significant threats to the environment and public health. Traditional gas sensors often lack high sensitivity and selectivity. To address this, our study uses first-principles density functional theory (DFT) to investigate CuO-SnS monolayers for improved gas sensor performance. The results show that CuO modification significantly enhances the adsorption capacity and selectivity of SnS monolayers for NO₂ and NO, with adsorption energies of −2.301 eV and −2.142 eV, respectively. Furthermore, CuO modification is insensitive to CO₂ adsorption, demonstrating excellent selectivity. Structural and electronic analyses reveal that CuO modification reduces the band gap of SnS monolayers from 1.465 eV to 0.635 eV, improving the electrical conductivity and electron transfer, thereby enhancing the gas adsorption sensitivity. Further analyses highlight significant electronic interactions and charge transfer mechanisms between CuO-SnS monolayers and NO₂ and SO₂ molecules, indicating strong orbital hybridization. In conclusion, this study provides a theoretical basis for developing high-performance gas sensors, showing that CuO-SnS monolayers have great potential for detecting toxic gases. Full article

Soil health is essential for sustainable agriculture, influencing ecosystem health and orchard productivity of plum orchards. Global challenges such as climate change and soil contamination threaten to affect fertility and food security, requiring sustainable practices. The study assessed the effect of different orchard management practices on soil quality and nutrient distribution in Prunus domestica L. orchard located on the Moldavian Plateau in northeastern Romania under temperate humid subtropical climate conditions. Two systems were analyzed: conventional (herbicide-based) and conservative (cover crop-based). Soil samples (0–20 cm and 20–40 cm) were analyzed for soil organic carbon (SOC), total nitrogen (N_t), available phosphorus (P), and potassium (K). Results showed that conservative management improved soil health by increasing SOC nutrient cycling, mainly through organic matter inputs. Compared to 2022, the effectiveness of phosphorus in the conservative management system significantly increased (by 6%) in 2023, while potassium content decreased (by 30%), suggesting potential nutrient competition or insufficient replenishment under organic practices. SOC levels remained stable, supporting long-term carbon inputs. Conventional management maintained phosphorus and potassium but showed lower SOC levels and higher risks of soil fertility depletion. Strong correlations between SOC and nutrient indicators emphasize the critical role of organic inputs in nutrient mobilization. The findings indicate that cover crops are essential for sustainable soil management by enhancing carbon sequestration and nutrient cycling, thereby supporting the long-term sustainability of agricultural systems. Full article

Our research introduces a framework that integrates edge computing, quantum transfer learning, and federated learning to revolutionize pain level assessment through ECG signal analysis. The primary focus lies in developing a robust, privacy-preserving system that accurately classifies pain levels (low, medium, and high) by leveraging the intricate relationship between pain perception and autonomic nervous system responses captured in ECG signals. At the heart of our methodology lies a signal processing approach that transforms one-dimensional ECG signals into rich, two-dimensional Continuous Wavelet Transform (CWT) images. These transformations capture both temporal and frequency characteristics of pain-induced cardiac variations, providing a comprehensive representation of autonomic nervous system responses to different pain intensities. Our framework processes these CWT images through a sophisticated quantum–classical hybrid architecture, where edge devices perform initial preprocessing and feature extraction while maintaining data privacy. The cornerstone of our system is a Quantum Convolutional Hybrid Neural Network (QCHNN) that harnesses quantum entanglement properties to enhance feature detection and classification robustness. This quantum-enhanced approach is seamlessly integrated into a federated learning framework, allowing distributed training across multiple healthcare facilities while preserving patient privacy through secure aggregation protocols. The QCHNN demonstrated remarkable performance, achieving a classification accuracy of 94.8% in pain level assessment, significantly outperforming traditional machine learning approaches. Full article

Global food supply chains face mounting vulnerabilities due to climate change and environmental variability, with particularly severe implications for import-dependent nations like Qatar, where over 90% of food supplies rely on international trade. This high import dependency creates unique challenges, including supply disruptions, price volatility, and food security risks, especially as climate variability increasingly affects major food-exporting regions. This study develops a multi-objective optimization framework to enhance the resilience of Qatar’s food import system by integrating economic, environmental, and crop water requirement considerations, modeled as a stochastic variable. The framework addresses both average performance and worst-case scenarios using stochastic and robust optimization approaches, evaluating trade partners for three key crops—tomatoes, onions, and cucumbers. Results identify optimal suppliers that minimize costs, environmental emissions, and water usage variability, with Turkey contributing 42.10% of total imports, Iran 13.76%, and the Netherlands 9.52%. The findings demonstrate that a diversified import strategy significantly reduces vulnerability to climate-induced disruptions and improves supply chain stability. This research provides actionable insights for policymakers, including; (1) optimal supplier diversification targets to balance risk and efficiency, (2) specific trade partner recommendations based on multiple sustainability criteria, and (3) quantitative frameworks for assessing import portfolio resilience. Full article

Lake areas across the Tibet Plateau have been taken as the major indicator of water resources changes. However, drivers behind spatiotemporal variations of lake areas over the Tibet Plateau have remained obscure. Selin Co Lake is the largest lake in the Qinghai–Tibet Plateau. Here, we delineate the Selin Co Lake area changes during the period of 1988–2023 based on Landsat remote sensing data. We also delved into causes behind the Selin Co Lake area changes from perspectives of glacier changes and tracing water vapor sources. We identified the persistently increasing lake area of Selin Co Lake. The Selin Co Lake area reached 2462.59 km² in 2023. We delineated the basin of Selin Co Lake and found a generally decreasing tendency of the main glaciers within the Selin Co basin. Specifically, the loss in the Geladandong Glacier area is 17.39 km² in total and the loss in the Jiagang Glacier area is 76.42 km². We found that the melting glaciers and precipitation within the Selin Co basin are the prime drivers behind the increasing the Selin Co Lake area. In the Selin Co basin, approximately 89.12% of the evaporation source of precipitation is propagated external to the Selin Co basin by the westerlies and the Indian monsoon. The internal hydrological circulation rate is 10.88%, while 30.61% of the moisture transportation is sourced from the ocean, and 69.39% is from the continental land. The moisture transportation from the ocean evaporation shows a significant increasing trend, which may contribute to the continued expansion of the Selin Co Lake area. Full article

In the context of sustainable development, water resources, energy, and carbon emissions are pivotal factors influencing the rational planning of economic development and the secure establishment of ecological barriers. As a core food production area, how can the Great River Basin balance the pressure on the “water–energy–carbon” system (WEC) to realize the coordinated development of “nature–society–economy”? Taking the Yellow River Basin in China as the research object, this paper explores the coupling characteristics and virtual transfer trends of WEC in the agricultural sector under the condition of mutual constraints. The results show the following: (1) On the dynamic coupling characteristics, W-E and E-C are strongly coupled with each other. The optimization of water resource allocation and the development of energy-saving water use technology make the W-E consumption show a downward trend, and the large-scale promotion of agricultural mechanization makes the E-C consumption show an upward trend. (2) On the spatial distribution of transfer, there is an obvious path dependence of virtual WEC transfer, showing a trend of transfer from less developed regions to developed regions, and the coupling strength decreases from developed regions to less developed regions. The assumption of producer responsibility serves to exacerbate the problem of inter-regional development imbalances. (3) According to the cross-sectoral analysis, water resources are in the center of sectoral interaction, and controlling the upstream sector of the resource supply will indirectly affect the synergistic relationship of WEC, and controlling the downstream sector of resource consumption will indirectly affect the constraint relationship of WEC. This study provides theoretical and methodological references for the Great River Basin to cope with the resource and environmental pressure brought by global climate change and the effective allocation of inter-regional resources. Full article

The pursuit of energy security has become one of the most important challenges facing modern societies worldwide. The increase in energy consumption and the need to promote sustainability puts pressure on power generation systems. In this context, renewable energy sources have become a favorable option to improve both energy security and sustainability while promoting the use of domestic energy sources. The supply chain is an optimized methodology that includes all necessary activities to bring a product to the final consumer. Traditionally applied in the manufacturing industry, recent evidence shows its successful implementation in various renewable energy sectors. In this work, a novel methodology based on a supply chain was designed to evaluate the feasibility of mini wind power systems in urban areas in an integrated and measurable manner. The main contribution lies in the integration of several different approaches, currently recognized as the most relevant factors for determining the viability of wind energy projects. A five-link supply chain model was proposed, which includes an evaluation of wind potential, supplier network, project technical assessment, customer distribution, and equipment final disposal. Specific metric indicators for each link were developed to evaluate technical, legislative, and social considerations. The methodology was applied in a case study in the city of Mexicali, Mexico. The findings show that although wind as a resource remains the most important factor, local government policies that promote the use of renewable energy, the supplier’s availability, qualified human resources, and spare parts are also of equivalent significance for the successful implementation of mini wind power systems. Full article

A low-voltage distribution substation is the last link before electricity transmission from the high-voltage grid to end-users. It is responsible for converting high-voltage electricity into low-voltage electricity suitable for domestic and commercial use and plays a central and critical role in the power system. The traditional modeling method is difficult to directly observe and solve the complete safety boundary expression in the high-dimensional state space, so the solution efficiency is greatly reduced. To address the above problems, this paper proposes a low-voltage distribution substation station safety domain (LVDS-SR) modeling method based on the high-dimensional safety domain definition method. In this paper, the concepts of safety work point, safety boundary, and safety domain are first defined. Then, the general mathematical model, edge points, and safety boundaries of the substation system are solved accurately by the high-dimensional safety domain definition and solution method, to obtain the safety domain model. The validity of the model and method is verified by arithmetic examples. Comparison with existing studies shows that the complete analytical formulation of the high-dimensional security domain is obtained for the first time in this paper, and the linearization method is used to improve the solution efficiency at the same time. This study provides a new analytical tool for the reliable and stable operation of low-voltage distribution substations, which has important theoretical and practical application value for the security assessment and optimization of power systems. Full article

In this research, we chose six indicators—soil conservation, water conservation, carbon sequestration, windbreak and sand fixation, biodiversity conservation, and forest recreation—to compute the forestland ecosystem service index for forestland within the study region, utilizing time series data. The outcomes reveal that the aggregate index of forestland ecosystem services exhibits a spatial distribution characterized by higher values in the northeastern part and lower values in the southwestern part, with an upward trend over time. Among these functions, windbreak and sand fixation, water conservation, carbon sequestration, and forest recreation all maintained relatively high growth rates. We selected 10 factors that are closely related to the natural environment and human activities and employed spatial principal component analysis to develop a comprehensive resistance surface. Based on the assessment results of forestland ecosystem functions, in conjunction with morphological spatial pattern analysis (MSPA) as well as landscape connectivity analysis, we optimized the method for identifying ecological source sites and extracted 38 ecological source sites. Subsequently, leveraging circuit theory, we extracted 91 ecological corridors and pinpointed 25 ecological nodes, ultimately constructing a forestland ecosystem security pattern (ESP) in the study area and proposing restoration strategies. Full article

In open-pit mining, autonomous trucks are essential for enhancing both safety and productivity. Object detection technology is critical to their smooth and secure operation, but training these models requires large amounts of high-quality annotated data representing various conditions. It is expensive and time-consuming to collect these data during open-pit mining due to the harsh environmental conditions. Simulation engines have emerged as an effective alternative, generating diverse labeled data to augment real-world datasets. However, discrepancies between simulated and real-world environments, often referred to as the Sim2Real domain shift, reduce model performance. This study addresses these challenges by presenting a novel semi-supervised domain adaptation for object detection (SSDA-OD) framework named Adamix, which is designed to reduce domain shift, enhance object detection, and minimize labeling costs. Adamix builds on a mean teacher architecture and introduces two key modules: progressive intermediate domain construction (PIDC) and warm-start adaptive pseudo-label (WSAPL). PIDC builds intermediate domains using a mixup strategy to reduce source domain bias and prevent overfitting, while WSAPL provides adaptive thresholds for pseudo-labeling, mitigating false and missed detections during training. When evaluated in a Sim2Real scenario, Adamix shows superior domain adaptation performance, achieving a higher mean average precision (mAP) compared with state-of-the-art methods, with 50% less labeled data required, achieved through active learning. The results demonstrate that Adamix significantly reduces dependence on costly real-world data collection, offering a more efficient solution for object detection in challenging open-pit mining environments. Full article

This study proposes a novel hybrid convolution and vision transformer model (ConvViT) designed to detect harmful insect species that adversely affect agricultural production and play a critical role in global food security. By utilizing a dataset comprising images of 15 distinct insect species, the suggested approach combines the strengths of traditional convolutional neural networks (CNNs) with vision transformer (ViT) architectures. This integration aims to capture local-level morphological features effectively while analyzing global spatial relationships more comprehensively. While the CNN structure excels at discerning fine morphological details of insects, the ViT’s self-attention mechanism enables a holistic evaluation of their overall configurations. Several data preprocessing steps were implemented to enhance the model’s performance, including data augmentation techniques and strategies to ensure class balance. In addition, hyperparameter optimization contributed to more stable and robust model training. Experimental results indicate that the ConvViT model outperforms commonly used benchmark architectures such as EfficientNetB0, DenseNet201, ResNet-50, VGG-16, and standalone ViT, achieving a classification accuracy of 93.61%. This hybrid approach improves accuracy and strengthens generalization capabilities, delivering steady performance during training and testing phases, thereby increasing its reliability for field applications. The findings highlight that the ConvViT model achieves high efficiency in pest detection by integrating local and global feature learning. Consequently, this scalable artificial intelligence solution can support sustainable agricultural practices by enabling the early and accurate identification of pests and reducing the need for intensive pesticide use. Full article