Search Results (2,696)

Accurate and physically consistent state prediction is essential for shipboard power systems (SPS) operating under dynamic conditions. However, purely data-driven models often exhibit degraded robustness and physically inconsistent outputs when exposed to transient disturbances or limited data coverage. To address these limitations, this paper proposes a physics-constrained hybrid prediction model that integrates a convolutional neural network–bidirectional long short-term memory (CNN–BiLSTM) architecture with wide residual connections (WRC) and a physics-constrained loss (PCL). The proposed modeling approach combines real operational measurement data with high-resolution simulation data to enhance data diversity and improve generalization capability. The CNN–BiLSTM structure captures nonlinear temporal dependencies, while the WRC preserves critical low-level transient electrical features during deep temporal modeling. In addition, multiple physical constraints, including power balance, voltage conversion relationships, and battery state-of-charge (SOC) dynamics, are incorporated into the training process to enforce physically consistent predictions. The model is validated using charging and discharging experiments on a laboratory-scale SPS under both steady-state and transient conditions. Comparative results demonstrate that the proposed approach achieves higher prediction accuracy, improved dynamic stability, and faster recovery following disturbances compared with conventional data-driven models. These results indicate that physics-constrained deep learning provides an effective and interpretable modeling framework for SPS state prediction, supporting digital twin-oriented monitoring and real-time prediction applications. Full article

Background: The Magnetic Resonance Imaging (MRI)-based tumor segmentation remains a challenging problem in medical imaging due to tumor heterogeneity, unpredictable morphological features, and the high complexity of calculations needed to implement it in clinical practice, putting it out of the scope of real-time applications. Although neural networks have significantly improved segmentation performance, they still struggle to capture morphological tumor features while maintaining computational efficiency. This work introduces Hybrid Adaptive Attention U-Net (HAAU-Net) framework, combining context-aware morphologically stable features and spatial channel attention to achieve high-quality tumor segmentation with less computational cost. Methods: The proposed HAAU-Net framework integrates multi-scale Adaptive Attention Blocks (AAB), Context-Aware Morphological Feature Module (CAMFM) and Spatial-Channel Hybrid Attention Mechanism (SCHAM). CAMFM is used to maintain the stability of morphological features by hierarchical aggregation and dynamic normalization of features. SCHAM enhances feature representation by modelling channels and spatial regions where the strongest feature are determined to use in segmentation. On the BRaTS 2022/2023 data, the proposed HAAU-Net is evaluated using four modalities including T1, T1GD, T2 and T2-FLAIR sequences. Results: The proposed model able to obtain 96.8% segmentation accuracy with a Dice coefficient of 0.89 on the entire tumor region, outperforming the alternative U-Net (0.83) and conventional CNN methods of segmentation (0.81). The proposed HAAU-Net architecture cuts the computational complexity of the standard deep learning models by 43% and still achieve real-time inference (28 FPS on a regular GPU). The hybrid model used to predict survival has a C-Index of 0.91 which is higher than the traditional SVM-based methods (0.72). Conclusions: Spatial-channel attention, combined with morphologically stable features, can be combined to allow clinically significant interpretability in attention maps. The proposed framework significantly improves segmentation performance while maintaining computational effeciency. This broad system has a serious potential of AI-enabled clinical decision support system and early prognostic diagnosis in neuro-oncology with practical deployment capability. Full article

Volatile organic compounds (VOCs, dominated by xylene, toluene, and benzene) and paint mist emissions from ship painting represent a major environmental and health concern, posing a critical bottleneck to the green transformation of the shipbuilding industry. To tackle this challenge, this study presents an integrated recovery system designed specifically for ship automatic-spraying robots. Guided by the synergistic principle of “air-curtain containment, multi-stage adsorption, and negative-pressure recovery,” the system features a modular design that ensures full compatibility with the robots’ spraying trajectory without operational interference. Core adsorption materials, namely glass fiber filter cotton and honeycomb activated carbon fiber, were selected to suit the high-humidity and high-pollutant-concentration environment typical of ship painting. An appropriately matched axial flow fan maintains stable negative pressure throughout the system. Furthermore, the design integrates an air curtain isolation subsystem and an automated control subsystem, enabling coordinated operation and real-time adjustment. Using ANSYS Fluent, geometric and flow field simulation models were established to analyze airflow distribution and pollutant adsorption behavior, which led to the optimization of key structural and material parameters. Field experiments conducted in shipyard environments demonstrated the system’s superior performance: it achieved a VOC removal efficiency of 88.4% and a paint mist capture efficiency of 85.7% under optimal working conditions, with a maximum simulated paint mist capture efficiency of 86.2%. The system maintained stable performance under complex vertical and overhead spraying conditions, with an efficiency attenuation of less than 1.5%, and its outlet emissions fully complied with the mandatory limits specified in the Emission Standard of Air Pollutants for the Shipbuilding Industry (GB 30981.2-2025). The relative error between experimental data and simulation results is less than 2%, confirming the reliability and practicality of the proposed system. This research provides an efficient and adaptable pollution control solution for green shipbuilding and offers valuable technical insights for the sustainable upgrading of automated painting processes in heavy industries. Full article

(1) Background: The scarcity and imbalance of real fault data significantly limit the development of data-driven fault diagnosis methods for liquid rocket engines (LREs), especially under few-shot conditions. (2) Methods: To address this issue, this study proposes a GAN-based fault data augmentation framework for multivariate LRE time-series signals and a hybrid diagnostic classifier combining convolutional neural networks (CNNs), bidirectional long short-term memory (BiLSTM), and multi-head attention (MHA). The GAN component is introduced to alleviate fault-data scarcity and class imbalance by generating additional fault-like samples, while the classifier is designed to capture local features, long-range temporal dependencies, and diagnostically informative temporal regions. (3) Results: A multidimensional evaluation based on temporal similarity, statistical consistency, and global distribution discrepancy indicates that the generated samples preserve important characteristics of the original signals under the current evaluation protocol. On the augmented LRE dataset, the proposed classifier achieved strong diagnostic performance. In addition, supplementary experiments on the public HIT aero-engine dataset further support the effectiveness of the classifier architecture, its component-wise contribution, and its behavior under imbalanced few-shot settings, while also demonstrating the value of uncertainty-aware prediction. (4) Conclusions: The results provide encouraging evidence that the proposed framework can improve LRE fault diagnosis under data-scarce conditions. However, the present findings should be interpreted within the scope of the available data and evaluation setting. More comprehensive generator-side ablation, broader external validation, and physics-oriented assessment of the generated signals are still needed before stronger conclusions can be made. Full article

Elevator fault diagnosis heavily relies on high-precision sensing of microscopic physical states. Although Micro-Electro-Mechanical System (MEMS) sensors can capture such subtle features, they are constrained by high-frequency data streams, environmental noise, and the semantic gap between raw sensor data and actionable maintenance decisions. This study proposes a collaborative edge–cloud intelligent diagnosis framework specifically designed for elevator systems. On the edge side, a lightweight temporal Transformer model, ELiTe-Transformer, was designed and deployed on the Jetson platform. This model enhances sensitivity to event-driven MEMS signals through an industrial positional encoding mechanism and by integrating linear attention and INT8 quantization techniques, achieving a real-time inference latency of 21.4 ms. On the cloud side, retrieval-augmented generation (RAG) technology was adopted to integrate physical features extracted at the edge with domain knowledge, generating interpretable diagnostic reports. The experimental results show that the overall accuracy of the system reaches 96.0%. The edge–cloud collaborative framework improves the accuracy of complex fault diagnosis to 92.5%, and the adoption of RAG reduces the report hallucination rate by 71.4%. This work effectively addresses the bottlenecks of MEMS perception in elevator fault diagnosis, forming a closed loop from micro-signal acquisition to high-level decision support. Full article

In recent years, DC-DC partial power converters (PPCs) have become increasingly important in fast-charging architectures for electric vehicles (EVs). Their key feature is that only a fraction of the energy delivered to the battery is processed by the PPC, while the rest is transferred directly, bypassing the conversion stage. This reduces DC-DC conversion losses and improves overall charging efficiency. However, the nonlinear dynamics of these converters can limit performance, especially with model-based controllers such as proportional–integral (PI) controllers. This paper proposes a data-driven control framework for EV fast-charging stations using a DC-DC PPC that is controlled by deep reinforcement learning (DRL). A value-based deep Q-network (DQN) directly selects switching actions and jointly regulates the partial-voltage and output current. The control problem is formulated as a discrete-time Markov decision process, and a two-stage transfer learning scheme ensures safe, efficient deployment. Firstly, the DQN agent is trained in a high-fidelity simulation and then fine-tuned with a small set of experimental data to capture parasitic and modeling errors. The controller is integrated into a constant-current–constant-voltage (CC-CV) charging algorithm and validated over a full charging cycle of a 60 kWh EV battery. The proposed control scheme exhibits a settling time of approximately 2 ms in response to current reference variations while maintaining steady-state errors below 2% in current regulation and below 1% in partial voltage regulation. Simulation results show that the proposed DRL controller has a small steady-state tracking error and improved robustness to reference changes compared with conventional PI and sliding mode controllers. The low computational cost of the trained DQN policy also enables real-time execution on embedded platforms for EV charging. Full article

Short-term traffic load prediction is a fundamental component of intelligent transportation systems (ITSs), supporting real-time monitoring, congestion mitigation, and adaptive traffic management in smart cities. Owing to the dynamic and nonlinear nature of urban traffic, identifying prediction models that align with real-world traffic dynamics remains a key challenge. This study presents a comparative evaluation of data-driven traffic load prediction models using high-resolution one-minute traffic data collected from a major urban roundabout in Jeddah, Saudi Arabia. The evaluated models include regression-based machine learning approaches and recurrent deep learning architectures, which are assessed under consistent preprocessing and evaluation conditions. Model performance is evaluated using standard error metrics and complemented by temporal and residual analyses to examine prediction behavior under different traffic regimes. The optimized GRU model achieved the best predictive accuracy with an RMSE of 149.12 veh/h, followed closely by the optimized LSTM model (RMSE = 150.85 veh/h). The results indicate that while conventional machine learning models can effectively capture overall traffic trends under relatively stable conditions, recurrent deep learning models demonstrate stronger capability in modeling nonlinear temporal dependencies and rapid traffic fluctuations when properly configured. In addition, a variability-based regime analysis was conducted to evaluate model robustness under different traffic demand dynamics, revealing that model performance advantages are context-dependent rather than universal. The findings highlight the importance of systematic comparative evaluation and data-driven model selection for developing reliable traffic prediction components in real-time ITS applications and sustainable urban mobility planning. Full article

The operation of the gas–steam–electricity multi-energy coupling system in iron and steel enterprises faces critical challenges: conflicts between energy efficiency and economic objectives, insufficient scheduling accuracy, and low energy utilization caused by source–load fluctuations. To address these issues, this paper proposes a hybrid scheduling model based on condition awareness and multi-objective optimization. The model integrates three key components. First, an energy fluctuation prediction technology based on working condition changes is developed. By acquiring real-time production signals and gas flow data, combined with a condition definition management module, it enables automatic identification and tracking of equipment operation status. A working condition sample curve superposition method is used to calculate energy medium imbalances, generating visual prediction curves for key parameters such as blast furnace, coke oven, and converter gas holder levels, achieving an average prediction accuracy of ≥95%. Second, a peak-shifting and valley-filling scheduling model for gas holders is designed, leveraging time-of-use electricity prices. During valley price periods, power purchases are increased and surplus gas is stored; during peak price periods, gas power generation is increased to reduce purchased electricity. A nonlinear model capturing the load–efficiency relationship of boilers and generators is established to dynamically optimize scheduling strategies. This reduces the proportion of peak hour power purchases by 10.3%, energy costs by 3.12%, and system energy consumption by 2.16%. Third, a multi-period and multi-medium energy optimization scheduling model is formulated as a mixed-integer nonlinear programming (MINLP) problem, with dual objectives of minimizing operating cost and energy consumption. Constraints include energy supply–demand balance, equipment operating limits, gas holder capacity, and generator ramp rates. The Pareto optimal solution set is obtained using the AUGMECON2 method and efficiently computed with the IPOPT solver. Application results demonstrate that the model achieves zero gas emissions, a dispatching instruction accuracy of 95%, and a 0.8% increase in the proportion of peak–valley-level self-generated power, outperforming comparable technologies. It provides technical support for the safe, efficient, and economic operation of multi-energy systems in iron and steel enterprises. Full article

This paper proposes a physics-informed hybrid digital CO₂ emission-control technology for maritime transport, designed for adaptive ship speed optimization along a predefined geographical route between two ports, discretized into quasi-stationary segments and evaluated under forecasted metocean conditions, subject to economic and regulatory constraints associated with maritime decarbonization. The framework integrates two exact optimization methods, Backtracking (BT) and Dynamic Programming (DP), with a reinforcement learning approach based on Proximal Policy Optimization (PPO), operating on a unified physical, economic, and regulatory modeling core. By reducing propulsion fuel demand, the system acts as an upstream CO₂ emission-control mechanism for ship propulsion. This operational stabilization of the engine load creates favourable boundary conditions for advanced combustion processes and reduces the volumetric flow of exhaust gas, thereby lowering the technical burden on potential post-combustion carbon capture systems. Segment-wise speed profiles are optimized subject to propulsion limits, Estimated Time of Arrival (ETA) feasibility, and regulatory constraints, including the Carbon Intensity Indicator (CII), the European Union Emissions Trading System (EU ETS) and FuelEU Maritime. The physics-based propulsion and energy model is validated using full-scale operational data from four real voyages of an oil/chemical tanker. A detailed case study on the Milazzo–Motril route demonstrates that adaptive speed optimization consistently outperforms conventional cruise operation. Exact optimization methods achieve voyage time reductions of approximately 10% and fuel and CO₂ emission reductions of about 9–10%. The reinforcement learning approach provides the best overall performance, reducing voyage time by approximately 15% and achieving fuel savings and CO₂ emission reductions of about 13%. At the route level, the Carbon Intensity Indicator is reduced by approximately 10% for the exact methods and by about 13% for PPO. Backtracking and Dynamic Programming converge to nearly identical globally optimal solutions within the discretized decision space, while PPO identifies solutions located on the most favourable region of the cost–time Pareto front. By benchmarking reinforcement learning against exact discrete solvers within a shared physics-informed structure, the proposed digital platform provides transparent validation of learning-based optimization and offers a scalable decision-support technology for pre-fixture evaluation of fixed-route voyages. The system enables quantitative assessment of CO₂ emissions, ETA feasibility, and regulatory exposure (CII, EU ETS, FuelEU Maritime penalties) prior to transport contracting, thereby supporting economically and environmentally informed operational decisions. Full article

Cr-based composite coatings with superior wear resistance are in growing demand for high-performance applications in the automotive, aerospace, and general manufacturing sectors. In this study, an Al10Cu alloy produced via powder metallurgy was coated with a chromium/nanodiamond (Cr/ND) composite layer using an electrodeposition process to enhance its tribological performance. The coatings were characterized using scanning electron microscopy, energy-dispersive X-ray spectroscopy, and X-ray diffraction. The resulting Cr/ND layer exhibited a uniform thickness of 73.5–76.2 μm and markedly improved surface hardness (809.4 HV), representing a 15-fold increase over the uncoated alloy (53.6 HV). Pin-on-disk tribological testing under dry sliding conditions showed complete elimination of detectable mass loss (0.00 mg vs. 0.55 mg for uncoated) within the measurement system resolution, indicating excellent resistance to both abrasive and adhesive wear. XRD analysis revealed the formation of a hexagonal close-packed Cr₂H phase with incorporated nanodiamond particles. To capture and predict the temporal evolution of the friction coefficient, a customized dual-layer long short-term memory neural network—optimized with a look-back window of 3 timesteps and ReLU-activated dense layers—was implemented. The model achieved superior predictive performance on the coated system, with validation and test R² values of 0.9973 and 0.9965, respectively, demonstrating enhanced modeling accuracy for surface-engineered materials. These findings demonstrate a significant advancement in wear protection for aluminum alloys and introduce a robust data-driven approach for real-time friction prediction in engineered surfaces. Full article

Embedded devices in modern power systems offer increased connectivity and remote reprogrammability/reconfigurability. These features along with interconnections between Information Technology (IT) and Operational Technology (OT) networks enable greater agility, reduced operator workload, and enhanced power system performance and capabilities, as well as expanding the cyber-attack surface. This increased cyber-attack surface, as well as increasingly complex, diverse, and potentially untrustworthy software/hardware supply chains, increases the need for robust real-time monitoring in power systems, and more generally in cyber–physical systems (CPS). We propose a novel framework for real-time monitoring and anomaly detection in CPS, specifically smart grid substations and SCADA systems. The proposed framework enables real-time signal temporal logic condition-based anomaly monitoring by processing raw captured packets from the communication network through a hierarchical semantic extraction and tag processing pipeline into a time series of semantic events and observations, that are then evaluated against expected temporal properties to detect and localize anomalies. We demonstrate the efficacy of our methodology on a hardware in the loop (HITL) testbed under several attack scenarios. The HITL testbed includes multiple physical power system devices (real-time automation controllers and relays) and simulated devices (Phasor Measurement Units—PMUs, relays, Phasor Data Concentrators—PDCs), all interfaced to a dynamic power system simulator. Full article

Rapid and accurate prediction of seismic response and fragility for bridge pile-group foundations (PGFs) is crucial for assessing seismic resilience. However, the high computational cost of traditional high-fidelity nonlinear analysis limits the application of probabilistic seismic risk analysis. To address this, an integrated deep learning framework is proposed that employs a unidirectional, multi-layer LSTM network for end-to-end prediction of structural responses directly from ground motions. The proposed model features two innovations. First, its multi-output capability enables simultaneous prediction of complete response time histories and peak values for key engineering demand parameters—bending moment, curvature, and pile cap displacement. Second, the network incorporates sliding time windows and residual connections to capture complex nonlinear soil–structure interaction. These predictions are integrated into a probabilistic seismic demand model to generate fragility curves. The framework is validated using a high-fidelity OpenSees model of a real bridge PGF subjected to 1000 ground motions. Results demonstrate the model’s excellent predictive accuracy: for peak bending moment, the mean predicted-to-actual ratio ranges from 0.97 to 1.03, with standard deviation below 0.12; the derived fragility curves show excellent agreement with benchmarks, achieving an average R² of 0.985 across four damage states. More importantly, the framework reduces the time for a complete fragility assessment (200 incremental dynamic analyses) from approximately 12 h to about 1 s—a 40,000× speed-up—making data-driven rapid and large-scale seismic risk assessment a reality. The proposed framework provides engineers with a practical design tool for rapidly evaluating alternative foundation configurations and informing seismic design decisions, thereby integrating advanced data-driven methods directly into the engineering design workflow. Full article

Tropical cyclones rank among the most destructive natural hazards globally, posing significant threats to coastal ecosystems and communities. Mangrove forests, renowned for their ecological importance and coastal protection services, are vulnerable to these disturbances, suffering structural damage, habitat loss, and disruption of vital ecosystem functions. Conventional field-based assessment methods often fall short in capturing the rapid and widespread impacts of cyclones, particularly in remote or cloud-obscured regions. This review aims to provide a comprehensive synthesis of remote sensing applications for monitoring cyclone-induced impacts on mangrove and coastal ecosystems worldwide. Through a systematic literature review of 74 peer-reviewed articles from 1990 to 2025, the study evaluates the utility of optical sensors, radar systems, and multi-sensor platforms in assessing inundation, vegetation damage, and ecosystem service loss. Key methodological advances such as time-series analysis, machine learning, and UAV-based validation are highlighted, alongside critical gaps including limited geographic coverage, weak validation practices, and minimal socio-economic integration. Notably, 75.4% of reviewed studies are concentrated in Asia, with Bangladesh and India alone accounting for 44.6% of the total literature, underscoring a pronounced geographic bias. The findings underscore the need for robust, near-real-time monitoring frameworks that combine satellite technologies with ground data and community engagement. Ultimately, the review advocates for an integrated, multi-sensor, and participatory approach to cyclone resilience, offering valuable insights for future research, disaster response planning, and sustainable mangrove management. Full article

Chronic Kidney Disease (CKD) affects over 850 million people globally, with early detection critical for effective intervention. We present DeepCKD-Net, a hybrid deep learning framework that synergistically integrates transformer architectures with gradient-boosting ensembles for multi-stage CKD prediction. Using a clinical dataset of 400 patients with 26 biomarker features from the UCI repository, our framework introduces three key innovations: (1) a hierarchical attention mechanism capturing complex inter-dependencies among clinical parameters, (2) an adaptive feature fusion module combining transformer-learned patterns with gradient-boosting decision boundaries, and (3) a confidence-aware ensemble strategy providing uncertainty quantification for clinical decision support. DeepCKD-Net achieves 98.7% accuracy and 0.993 AUC, surpassing state-of-the-art methods by 4.2% while maintaining 16.8 ms inference time suitable for real-time clinical deployment. Integrated SHAP analysis provides interpretable predictions, with serum creatinine (SHAP value: 0.342) and blood urea (0.287) identified as top predictive biomarkers, aligning with established clinical knowledge. The framework demonstrates robust performance under realistic clinical conditions, maintaining >90% accuracy with 20% missing data. Our contributions advance AI-driven nephrology diagnostics by providing a deployable, interpretable, and clinically validated solution for early CKD detection. Full article

To meet the real-time computational requirements of active suspension control systems, this study shifts from complex microscopic physical equations to a direct nonlinear functional mapping between the relative motion states (displacement and velocity) and the output force of air springs. This approach aims to preserve critical nonlinear hysteresis characteristics while significantly reducing the computational overhead. A progressive modeling strategy is implemented to characterize these complex behaviors. Initially, polynomial fitting is employed to identify key input features; however, its limited capacity to capture intricate nonlinearities necessitates more advanced methods. Subsequently, standard Feedforward Neural Networks (FNNs) are explored for their nonlinear mapping capabilities, yet their inherent “black-box” nature often leads to convergence difficulties and restricted generalization. To address these issues, a Physics-Informed Neural Network (PINN) architecture is introduced, embedding physical governing equations as regularization constraints within the loss function to integrate data-driven flexibility with mathematical rigor. Recognizing that conventional PINNs often encounter convergence challenges due to conflicts between PDE constraints and data-driven loss terms, this research develops a Physics-Embedded Hierarchical Network (PEHN). By deriving specialized PDE constraints tailored to air spring dynamics and designing a hierarchical architecture aligned with these physical requirements, the PEHN effectively balances physical priors with experimental data. Experimental results demonstrate that, compared to the baseline models, the proposed PEHN exhibits stronger stability and superior accuracy in capturing the complex nonlinearities of air spring dynamics. Full article