Search Results (8,313)

The rapid expansion of Internet of Things (IoT)-enabled smart grids has intensified the need for reliable fault detection and autonomous self-healing under non-stationary operating conditions characterized by frequent concept drift. To address the limitations of static and single-strategy adaptive models, this paper proposes H-CLAS, a novel Hybrid Continual Learning for Adaptive Self-healing framework that unifies regularization-based, memory-based, architectural, and meta-learning strategies within a single adaptive pipeline. The framework integrates convolutional neural networks (CNNs) for fault detection, graph neural networks for topology-aware fault localization, reinforcement learning for self-healing control, and a hybrid drift detection mechanism combining ADWIN and Page–Hinkley tests. Continual adaptation is achieved through the synergistic use of Elastic Weight Consolidation, memory-augmented replay, progressive neural network expansion, and Model-Agnostic Meta-Learning for rapid adaptation to emerging drifts. Extensive experiments conducted on the Smart City Air Quality and Network Intrusion Detection Dataset (NSL-KDD) demonstrate that H-CLAS achieves accuracy improvements of 12–15% over baseline methods, reduces false positives by over 50%, and enables 2–3× faster recovery after drift events. By enhancing resilience, reliability, and autonomy in critical IoT-driven infrastructures, the proposed framework contributes to improved grid stability, reduced downtime, and safer, more sustainable energy and urban monitoring systems, thereby providing significant societal and environmental benefits. Full article

Very Low Earth Orbit (VLEO) satellites, operating at altitudes below 450 km, provide tremendous potential in the domain of remote sensing. Their proximity to Earth offers high resolution, low latency, and rapid revisit rates, allowing continuous monitoring of dynamic systems and real-time delivery of vertically integrated earth observation products. Nonetheless, the application of VLEO is not yet fully realized due to numerous complexities associated with VLEO satellite development, considering atmospheric drag, short satellite lifetimes, and social, political, and legal regulatory fragmentation. This paper reviews the recent technological developments supporting sustainable VLEO operations with regards to aerodynamic satellite design, atomic oxygen barriers, and atmospheric-breathing electric propulsion (ABEP). Furthermore, the paper provides an overview of the identification of regulatory and economic barriers that extort additional costs for VLEO ranging from frequency band allocation and space traffic management to life-cycle cost and uncertain commercial demand opportunities. Nevertheless, the commercial potential of VLEO operations is widely acknowledged, and estimated to lead to an economic turnover in the order of 1.5 B USD in the next decade. Learning from the literature and prominent past experiences such as the DISCOVERER and CORONA programs, the study identifies key gaps and proposes a roadmap to sustainable VLEO development. The proposed framework emphasizes modular and serviceable satellite platforms, hybrid propulsion systems, and globally harmonized governance in space. Ultimately, public–private partnerships and synergies across sectors will determine whether VLEO systems become part of the broader space infrastructure unlocking new capabilities for near-Earth services, environmental monitoring, and commercial innovation at the edge of space. Full article

With the large-scale integration of power electronic converters, non-linear loads, and renewable energy generation, voltage and current waveform distortion in modern power systems has become increasingly severe, making harmonic resonance amplification and non-stationary distortion more prominent. Accurate and robust harmonic-level prediction and detection have become essential foundations for power quality monitoring and operational protection. However, traditional harmonic analysis methods remain highly dependent on pre-designed time–frequency transformations and manual feature extraction. They are sensitive to noise interference and operational variations, often exhibiting performance degradation under complex operating conditions. To address these challenges, a Unified Physics-Transformer-based harmonic detection scheme is proposed to accurately forecast harmonic levels in offshore wind farms (OWFs). This framework utilizes real-world wind speed data from Bozcaada, Turkey, to drive a high-fidelity electromagnetic transient simulation, constructing a benchmark dataset without reliance on generative data expansion. The proposed model features a Feature Tokenizer to project continuous physical quantities (e.g., wind speed, active power) into high-dimensional latent spaces and employs a Multi-Head Self-Attention mechanism to explicitly capture the complex, non-linear couplings between meteorological inputs and electrical states. Crucially, a Multi-Task Learning (MTL) strategy is implemented to simultaneously regress the Total Harmonic Distortion (THD) and the characteristic 5th Harmonic (H5), effectively leveraging shared representations to improve generalization. Comparative experiments with Random Forest, LSTM, and GRU systematically evaluate the predictive performance using metrics such as root mean square error (RMSE) and mean absolute percentage error (MAPE). Results demonstrate that the Physics-Transformer significantly outperforms baseline methods in prediction accuracy, robustness to operational variations, and the ability to capture transient resonance events. This study provides a data-efficient, high-precision approach for harmonic forecasting, offering valuable insights for future renewable grid integration and stability analysis. Full article

Mobile authentication is crucial for device security; however, conventional techniques such as PINs and swipe patterns are susceptible to social engineering attacks. This work explores the integration of touch stroke analysis and Explainable AI (xAI) to address these vulnerabilities. Unlike static methods that require intervention at specific intervals, continuous authentication offers dynamic security by utilizing distinct user touch dynamics. This study aggregates touch stroke data from 150 participants to create comprehensive user profiles, incorporating novel biometric features such as mid-stroke pressure and mid-stroke area. These profiles are analyzed using machine learning methods, where the Random Tree classifier achieved the highest accuracy of 97.07%. To enhance interpretability and user trust, xAI methods such as SHAP and LIME are employed to provide transparency into the models’ decision-making processes, demonstrating how integrating touch stroke dynamics with xAI produces a visible, trustworthy, and continuous authentication system. Full article

The widespread adoption of electric vehicles (EVs) has posed significant challenges to the security of distribution grid loads. To address issues such as increased grid load fluctuations, rising user charging costs, and rapid load surges around midnight caused by uncoordinated nighttime charging of household electric vehicles in communities, this paper first models electric vehicle charging behavior as a Markov Decision Process (MDP). By improving the state-space sampling mechanism, a continuous space mapping and a priority mechanism are designed to transform the charging scheduling problem into a continuous decision-making framework while optimizing the dynamic adjustment between state and action spaces. On this basis, to achieve synergistic load forecasting and charging scheduling decisions, a forecast-augmented deep reinforcement learning method integrating Gated Recurrent Unit and Twin Delayed Deep Deterministic Policy Gradient (GRU-TD3) is proposed. This method constructs a multi-objective reward function that comprehensively considers time-of-use electricity pricing, load stability, and user demands. The method also applies a single-objective pre-training phase and a model-specific importance-sampling strategy to improve learning efficiency and policy stability. Its effectiveness is verified through extensive comparative and ablation validation. The results show that our method outperforms several benchmarks. Specifically, compared to the Deep Deterministic Policy Gradient (DDPG) and Particle Swarm Optimization (PSO) algorithms, it reduces user costs by 11.7% and the load standard deviation by 12.9%. In contrast to uncoordinated charging strategies, it achieves a 42.5% reduction in user costs and a 20.3% decrease in load standard deviation. Moreover, relative to single-objective cost optimization approaches, the proposed algorithm effectively suppresses short-term load growth rates and mitigates the “midnight peak” phenomenon. Full article

Continuous missing data is a prevalent challenge in long-term air quality monitoring, undermining the reliability of public health protection and sustainable urban development. In this paper, we propose ConFill, a novel contrastive learning-based framework for reconstructing continuous missing data in air quality time series. By leveraging temporal continuity as a supervisory signal, our method constructs positive sample pairs from adjacent subsequences and negative pairs from distant and shuffled segments. Through contrastive learning, the model learns robust representations that preserve intrinsic temporal dynamics, and enable accurate imputation of continuous missing segments. A novel data augmentation strategy is proposed, to integrate noise injection, subsequence masking, and time warping to enhance the diversity and representativeness of training samples. Extensive experiments are conducted on a large scale real-world dataset comprising multi-pollutant observations from 209 monitoring stations across China over a three-year period. Results show that ConFill outperforms baseline imputation methods under various missing scenarios, especially in reconstructing long consecutive gaps. Ablation studies confirm the effectiveness of both the contrastive learning module and the proposed augmentation technique. Full article

Intrinsic motivation provides a principled mechanism for driving exploration in reinforcement learning when external rewards are sparse or delayed. A central challenge, however, lies in defining meaningful novelty signals in high-dimensional and combinatorial state spaces, where observation-level density estimation and prediction-error heuristics often become unreliable. In this work, we propose an information-theoretic framework for intrinsically motivated reinforcement learning grounded in the Information Bottleneck principle. Our approach learns compact latent state representations by explicitly balancing the compression of observations and the preservation of predictive information about future state transitions. Within this bottlenecked latent space, intrinsic rewards are defined through information-theoretic quantities that characterize the novelty of state–action transitions in terms of mutual information, rather than raw observation dissimilarity. To enable scalable estimation in continuous and high-dimensional settings, we employ neural mutual information estimators that avoid explicit density modeling and contrastive objectives based on the construction of positive–negative pairs. We evaluate the proposed method on two representative combinatorial routing problems, the Travelling Salesman Problem and the Split Delivery Vehicle Routing Problem, formulated as Markov decision processes with sparse terminal rewards. These problems serve as controlled testbeds for studying exploration and representation learning under long-horizon decision making. Experimental results demonstrate that the proposed information bottleneck-driven intrinsic motivation improves exploration efficiency, training stability, and solution quality compared to standard reinforcement learning baselines. Full article

Current ML lifecycle frameworks provide limited support for continuous stakeholder alignment and infrastructure evolution, particularly in sensor-based AI systems. We present CAPTURE, a seven-phase framework (Consult, Articulate, Protocol, Terraform, Utilize, Reify, Evolve) that integrates stakeholder-centered requirements engineering with MLOps practices to address these gaps. framework (Consult, Articulate, Protocol, Terraform, Utilize, Reify, Evolve) that integrates stakeholder-centered requirements engineering with MLOps practices to address these gaps. The framework was synthesized from four established standards (ISO/IEC 22989, ISO 9241-210, CRISP-ML(Q), SE4ML) and validated through a longitudinal five-year case study of a psychomotor skill learning system alongside semi-structured interviews with ten domain experts. The evaluation demonstrates that CAPTURE supports governance of iterative development and strategic evolution through explicit decision gates. Expert assessments confirm the necessity of the intermediate stakeholder-alignment layer and substantiate the participatory modeling approach. By connecting technical MLOps with human-centered design, CAPTURE reduces the risk that sensor-based AI systems become ungoverned, non-compliant, or misaligned with user
needs over time. Full article

Rapid urbanisation in coastal megacities intensifies coupled climate and water-related challenges, including heat stress, ventilation deficits, and increasing sensitivity to hydro-climatic extremes. Urban morphology plays a critical role in regulating these climate–water interactions by shaping airflow, surface heat exchange, and the spatial organisation of green–blue infrastructures. This study develops a Local Climate Zone (LCZ)-conditioned generative modelling framework based on a Conditional Pix2Pix Generative Adversarial Network, using paired LCZ classification maps and urban morphology data derived from six representative cities in the Guangdong–Hong Kong–Macao Greater Bay Area: Guangzhou, Shenzhen, Hong Kong, Macao, Zhuhai, and Dongguan. By integrating remote sensing–derived LCZ classifications with urban morphology data, the proposed framework learns spatial patterns associated with key morphology-related predictors, including building density and compactness, height-related structural intensity, open-space distribution, and the continuity of green–blue and ventilation corridors. The model demonstrates robust performance (SSIM = 0.74, R² = 0.81, PSNR = 15.3 dB) and strong cross-city transferability, accurately reproducing density transitions, ventilation corridors, and green–blue spatial structures relevant to coastal climate and water adaptation. The results highlight the potential of LCZ-informed generative modelling as a scalable decision-support tool for climate–water adaptive urban planning, enabling rapid exploration of morphology configurations that support heat mitigation, ventilation enhancement, and resilient coastal transformation. Full article

The rapid integration of unmanned aerial vehicles (UAVs) into next-generation wireless systems demands seamless and reliable handover (HO) mechanisms to ensure continuous connectivity. However, frequent topology changes, high mobility, and dynamic channel variations make traditional HO schemes inadequate for UAV-assisted 6G networks. This paper presents a comprehensive review of existing HO optimization studies, emphasizing artificial intelligence (AI) and machine learning (ML) approaches as enablers of intelligent mobility management. The surveyed works are categorized into three main scenarios: non-UAV HOs, UAVs acting as aerial base stations, and UAVs operating as user equipment, each examined under traditional rule-based and AI/ML-based paradigms. Comparative insights reveal that while conventional methods remain effective for static or low-mobility environments, AI- and ML-driven approaches significantly enhance adaptability, prediction accuracy, and overall network robustness. Emerging techniques such as deep reinforcement learning and federated learning (FL) demonstrate strong potential for proactive, scalable, and energy-efficient HO decisions in future 6G ecosystems. The paper concludes by outlining key open issues and identifying future directions toward hybrid, distributed, and context-aware learning frameworks for resilient UAV-enabled HO management. Full article

The pygmy hippopotamus (Choeropsis liberiensis, Morton, 1849) is classified as Endangered by the International Union for the Conservation of Nature (IUCN). Compared to other large, threatened mammals, this species remains relatively understudied and new findings indicate potential welfare concerns, emphasizing the need for further research on the species welfare in zoological institutions. One approach to improving welfare in captivity is through environmental enrichment. This study investigated the effects of olfactory enrichment on three individual pygmy hippopotamuses through behavioral analysis and heat-map visualization. Using continuous focal sampling, several behaviors were influenced by the stimuli, with results showing a general decrease in inactivity and an increase in environmental engagement and interaction, particularly through scenting behavior. To further enhance behavioral quantification, machine learning techniques were applied to video data, comparing manual and automated behavior classification using the pose estimation program SLEAP. Four behaviors Standing, Locomotion, Feeding/Foraging, and Lying Down were compared. A confusion matrix, time budgets, and Kendall’s Coefficient of Concordance (W) were used to assess agreement between methods. The results showed a strong and moderate agreement between manual and automated annotations, for the female and calf, respectively. This demonstrates the potential of automation to complement behavioral observations in future welfare monitoring. Full article

The objective of the present study was to apply supervised Machine Learning to predict severe complications after equine orchiectomy. A dataset of 612 cases of orchiectomies in stallions was used for the development of a computational model, among which in 8.5% of cases severe complications (colic, continued stallion-like behaviour, evisceration, funiculitis, haemorrhage, and scrotal infection) were diagnosed post-orchiectomy. Three supervised Machine Learning tools were employed: Logistic Regression (12 different models evaluated), Random Forest (64 models), and Gradient Boosting (8 models). For the prediction of the development of severe complications post-orchiectomy, Logistic Regression was the tool that produced the best discrimination measures, where accuracy, precision and recall were 0.9134, 0.8391, and 0.9133, respectively. The results of the analysis for SHapley Additive exPlanations values for the impact of the independent variables in the prediction of the development of complications indicated that (a) the age of the horse and (b) the surgical technique employed were the two variables that mostly influenced the prediction outcome, findings that were unambiguous in the models developed by any Machine Learning tool. The findings of this study indicate that computational models could be used as adjunct tools to support clinical decisions in the peri-operative management of horses. Full article

Rice is a vital staple food crop worldwide and also one of the major sources of greenhouse gas (GHG) emissions, generating substantial methane (CH₄) and nitrous oxide (N₂O). As one of the key management practices for rice production, the GHG mitigation potential of water management has attracted extensive attention, whereas its global scalability remains to be further investigated. Based on 15,458 global observations of field experimental data, we employed advanced machine learning methods to quantify the GHGs and soil carbon sequestration of global rice systems around 2020. Furthermore, we identified the optimal spatial distribution of GHG mitigation for five rice water management practices (continuous flooding (CF), flooding–midseason drainage–reflooding (FDF), alternate wetting and drying irrigation (AWD), flooding–midseason drainage–intermittent irrigation (FDI), and rainfed cultivation (RF)) through scenario simulation, under the premise of no yield reduction. The results of machine learning simulation showed that optimizing water management could reduce global rice greenhouse gas emissions by 39.17%, equivalent to 340.46 Mt CO₂ eq, while increasing rice yields by 3.55%. This study provides valuable insights for the optimization of agricultural infrastructure and the realization of agricultural sustainable development. Full article

Exposure to Particulate Matter (PM) is linked to respiratory and cardiovascular diseases, certain types of cancer, and accounts for approximately seven million premature deaths globally. While governments and organizations have implemented various strategies for Air Quality (AQ) such as the deployment of Air Quality Monitoring Networks (AQMN), these networks often suffer from limited spatial coverage and involve high installation and maintenance costs. Consequently, the implementation of networks based on Low-Cost Sensors (LCS) has emerged as a viable alternative. Nevertheless, LCS systems have certain drawbacks, such as lower reading precision, which can be mitigated through specific calibration models and methods. This paper presents the results and conclusions derived from simultaneous PM₁₀ and PM_2.5 monitoring comparisons between LCS nodes and a T640X reference sensor. Additionally, Relative Humidity (RH), temperature, and absorption flow measurements were collected via an Automet meteorological station. The monitoring equipment was installed at the Faculty of Environment of the Universidad Distrital in Bogotá. The LCS calibration process began with data preprocessing, which involved filtering, segmentation, and the application of FastDTW. Subsequently, calibration was performed using a variety of models, including two statistical approaches, three Machine Learning algorithms, and one Deep Learning model. The findings highlight the critical importance of applying FastDTW during preprocessing and the necessity of incorporating RH, temperature, and absorption flow factors to enhance accuracy. Furthermore, the study concludes that Random Forest and XGBoost offered the highest performance among the methods evaluated. While satellites map city-wide patterns and MAX-DOAS enables hourly source attribution, our calibrated LCS network supplies continuous, street-scale data at low CAPEX/OPEX—forming a practical backbone for sustained micro-scale monitoring in Bogotá. Full article

Machine learning (ML)-powered vision for robotic inspection has accelerated with smart manufacturing, enabling automated defect detection and classification and real-time process optimization. This review provides insight into the current landscape and state-of-the-art practices in smart manufacturing quality control (QC). More than 50 studies spanning across automotive, aerospace, assembly, and general manufacturing sectors demonstrate that ML-powered vision is technically viable for robotic inspection in manufacturing. The accuracy of defect detection and classification frequently exceeds 95%, with some vision systems achieving 98–100% accuracy in controlled environments. The vision systems use predominantly self-designed convolutional neural network (CNN) architectures, YOLO variants, or traditional ML vision models. However, 77% of implementations remain at the prototype or pilot scale, revealing systematic deployment barriers. A discussion is provided to address the specifics of the vision systems and the challenges that these technologies continue to face. Finally, recommendations for future directions in ML-powered vision for robotic inspection in manufacturing are provided. Full article