Search Results (6,763)

Phase change material (PCM)-based latent heat storage (LHS) systems help address the mismatch between renewable energy supply and thermal demand. However, their practical implementation is constrained by the strongly nonlinear and multiphysics nature of phase change, which makes high-fidelity simulations and real-time applications computationally expensive. This review examines mathematical reduced-order modeling (ROM) as an effective strategy to overcome this limitation by combining physics-based simplifications, projection methods, interpolation techniques, and data-driven models for PCM-based LHS systems. While physical simplifications (such as dimensional reduction and effective property approximations) represent an important first layer of model reduction, the primary focus of this work is on the mathematical ROM methodologies that operate on the governing equations after such physical simplifications have been applied. The review covers approaches including two-temperature non-equilibrium and analytical thermal-resistance models, Proper Orthogonal Decomposition (POD), CFD-derived look-up tables, kriging and ε-NTU grey/black-box metamodels, and machine-learning methods such as artificial neural networks and gradient-boosted regressors trained from CFD data. These ROM techniques have been applied to packed beds, PCM-integrated heat exchangers, finned enclosures, triplex-tube systems, and solar thermal components, achieving speed-ups from tens to over 80,000 times faster than full CFD simulations while maintaining prediction errors typically below 5% or within sub-Kelvin temperature deviations. A critical comparative analysis exposes the fundamental trade-off between interpretability, data dependence, and computational efficiency, leading to a practical decision-making framework that guides method selection for specific applications such as design optimization, real-time control, and system-level simulation. Remaining challenges—including accurate representation of phase change nonlinearity, moving phase boundaries, multi-timescale dynamics, generalization across geometries, experimental validation, and integration into industrial workflows—motivate a structured roadmap for future hybrid physics–machine learning developments, standardized validation protocols, and pathways toward industrial deployment. Full article

Artificial Intelligence (AI), particularly Machine Learning (ML) and Deep Learning (DL), is transforming the modeling of complex spatiotemporal urban processes such as urban growth, sprawl, shrinkage, redevelopment, and Land Use/Land Cover Change (LULCC). However, despite rapid methodological innovation, applications remain fragmented, and there is limited synthesis of how AI-based models complement, extend, or supersede conventional approaches. This study addresses this gap through a systematic review of 6356 records, from which 120 articles were selected for detailed analysis. It investigates: (i) how ML/DL techniques are embedded within spatiotemporal modeling frameworks; (ii) their use in simulating urbanization dynamics and land-use (LU) transitions; (iii) methodological and performance gains relative to traditional statistical and rule-based models; and (iv) emerging research frontiers and limitations. The review shows that LULCC dominates current applications, with Artificial Neural Networks (ANNs) as the most prevalent ML method, increasingly complemented by DL architectures. Across cases, AI is primarily used to learn non-linear transition dynamics, represent spatial and temporal dependencies, identify influential drivers, and improve classification performance and computational efficiency. Building on these insights, the paper synthesizes the roles of AI in spatiotemporal urban modeling and outlines forward-looking research directions to support more robust, transparent, and policy-relevant applications for urban sustainability. Full article

The accurate and explainable prediction of carbon emissions is crucial for the efficient operation of hybrid and electrified transportation systems and their integration with energy grids. An Explainable Spatio-Temporal Graph Neural Network (EST-GNN) is proposed for highly precise CO₂ emission forecasting using Lévy Flight-guided Optuna optimization. By modelling vehicles and their operational characteristics as nodes in a dynamic graph, the proposed framework can jointly learn timing and spatial correlations while sustaining interpretability. The accuracy of the EST-GNN model is compared with models based on one-hot encoded features, SMOTE-enhanced datasets, and ensemble regressors. Using a real-world dataset of 7385 vehicle registrations with 12 predictive features experiments are conducted. When applied the EST-GNN model outperformed all baseline and traditional models achieving the highest reliability (R² = 0.98754) while solving competitive error metrics (RMSE = 6.55, MAE = 2.556). There is strong indication that reasonable machine learning (ML) models can be used accurately to confirm their suitability for resource-prevented and real-time applications, while predictable ML techniques have relatively low reliability. The optimal solution ensures scalability, robustness, and independence of the deployment environment. The distribution analysis of best performing models develops the ability of EST-GNN, which accounts for the largest proportion of best results across evaluation metrics. To achieve superior predictive accuracy, graph-based learning, explainability, and advanced hyperparameter optimization are combined. EST-GNN provides a powerful tool for analyzing fleet emission levels, making energy-aware decisions, and planning sustainable transportation, while ML models continue to be a useful complement for deployment states with high computation costs and quick responses. Full article

Intelligent bearing fault diagnosis is essential for maintaining the reliability of rotating machinery. However, deploying deep learning models in industrial environments is often constrained by a lack of labeled data, environmental noise, and strict hardware limits. To address these connected challenges, this paper proposes 1D-CNN-Trans, a flexible and resource-efficient hybrid framework. Designed for supervised diagnosis with restricted data, the configurable model combines a compact one-dimensional convolutional neural network (1D-CNN) for local feature extraction, a Transformer encoder for capturing long-range temporal dependencies, and an optional squeeze-and-excitation (SE) module for channel recalibration under favorable conditions. The method is evaluated on two standard mechanical benchmarks under limited sample conditions, controlled additive white Gaussian noise (AWGN), and dynamic non-stationary interference. Experimental results indicate that 1D-CNN-Trans shows improved robustness under interference compared to selected baselines, notably improving accuracy against a standard CNN backbone. Furthermore, findings indicate that while the Transformer ensures noise robustness, channel recalibration (via SE) introduces optimization instability under extreme sparsity and noise. Consequently, we reposition the architecture as a configurable framework where recalibration is conditionally activated. Finally, theoretical complexity analysis is provided to validate the model’s low computational burden, indicating its general feasibility for resource-constrained scenarios. Full article

Accurate measurement of ocean wave parameters is paramount for offshore engineering design and marine environmental monitoring. Distributed pressure sensing technology provides a robust data foundation for analyzing the spatio-temporal characteristics of wave fields through synchronized observations at multiple stations. However, multi-sensor data exhibit high-dimensional spatio-temporal coupling, posing significant challenges for traditional single-point signal processing methods in capturing the topological associations between measurement sites. To address these limitations, this study develops a framework for spatio-temporal feature modeling and wave height reconstruction based on Graph Neural Networks (GNNs). The proposed framework integrates the spatial configuration of sensor arrays with graph-theoretic topological representations. By fusing geometric distances and signal correlations, an adaptive adjacency matrix is constructed to establish a dynamically adjustable graph structure. On the feature extraction level, a spatio-temporal fusion method combining multi-scale graph convolutions and gated temporal modeling is proposed. The experimental results obtained on the Blancs Sablons Bay multi-sensor dataset demonstrate that the proposed method significantly outperforms traditional approaches, achieving lower prediction errors and validating the effectiveness of graph-structured modeling in distributed wave sensing. Full article

Accurate air quality forecasting requires the effective integration of heterogeneous data sources that vary in spatial coverage, temporal resolution, and sensing reliability. This paper presents a unified spatio-temporal data processing framework designed to support multi-source air quality forecasting by jointly leveraging regulatory monitoring stations of the Central Pollution Control Board (CPCB) as reference-grade anchors and community-driven Internet of Things (IoT) sensing platforms for spatial densification. The proposed end-to-end workflow addresses key challenges associated with heterogeneity, data quality, and interoperability through systematic schema harmonization, multi-stage data cleaning, and robust missing data imputation using a Robocentric Iterated Extended Kalman Filter (RIEKF). The processed data are temporally aligned to a uniform sampling grid and enriched with spatial descriptors, including geospatial coordinates, administrative boundaries, and proximity-based emission features. These enriched observations are subsequently fused into a unified spatio-temporal representation that captures both spatial dependencies and temporal dynamics across the sensor network. Dynamic graphs constructed from this representation are processed using a Mobility-Aware Peripheral-Enhanced Graph Neural Network to forecast pollutant concentrations and generate categorical air quality indices. The framework is evaluated using regression metrics reported as RMSE/MAE in µg/m³ and MAPE in %, together with standard AQI classification metrics, demonstrating its effectiveness in improving predictive accuracy and robustness for real-world air quality forecasting applications. Full article

Credit risk prediction is central to financial stability and regulatory compliance, guiding lending decisions and portfolio risk management. While traditional approaches such as logistic regression and tree-based models have long been the industry standard, recent advances in deep learning (DL) have introduced architectures capable of capturing complex nonlinearities, temporal dynamics, and relational dependencies in borrower data. This study provides a comprehensive review of DL methods applied to credit risk prediction, covering multi-layer perceptron, recurrent and convolutional neural networks, transformer, and graph neural networks. We examine benchmark and large-scale datasets, highlight peer-reviewed applications across corporate, consumer, and peer-to-peer lending, and evaluate the benefits of DL relative to classical machine learning. In addition, we critically assess key challenges and identify emerging opportunities. By synthesising methods, applications, and open challenges, this paper offers a roadmap for advancing trustworthy deep learning in credit risk modelling and bridging the gap between academic research and industry deployment. Full article

This paper proposes a novel adaptive sliding mode control strategy for bionic aerial manipulators performing canopy-sampling tasks. Specifically, an adaptive neural barrier function-based fast terminal sliding mode control (BFASMC-NN) scheme is developed to address the joint-space trajectory tracking problem by integrating fast continuous nonsingular terminal sliding mode control (FNTSMC), neural networks (NNs), and barrier functions (BFs). The aerial manipulator is modeled as a rootless system, and its kinematic and dynamic characteristics are analyzed separately. Radial basis function neural networks (RBF-NNs) are introduced to approximate lumped disturbances, while BFs are incorporated to mitigate the effects of joint input saturation. Meanwhile, FNTSMC is employed to guarantee finite-time convergence of the system states. The stability of the closed-loop system is rigorously proven based on Lyapunov stability theory. Two simulation studies are conducted to validate the proposed method, and the results demonstrate that it achieves stronger disturbance rejection capability, faster convergence, and higher tracking accuracy than existing approaches. Full article

This paper presents an adaptive optimal consensus tracking control scheme for canonical nonlinear multi-agent systems (MASs) with unknown dynamics, employing an actor–critic reinforcement learning (RL) framework. The scheme integrates a sliding mode mechanism to suppress tracking errors and ensure consensus tracking between the followers and the leader. Additionally, optimal control is designed to find a Nash equilibrium in a graphical game. To address the intractability of obtaining an analytical solution for the coupled Hamilton–Jacobi–Bellman (HJB) equation, a policy iteration algorithm is utilized. Within this algorithm, a critic neural network (NN) approximates the gradient of the optimal value function, while an actor NN approximates the optimal control policy. Together, these networks form a compact actor–critic (AC) architecture that achieves optimal consensus tracking. Furthermore, the proposed method guarantees the boundedness of all closed-loop signals while ensuring consensus tracking. Finally, two simulations are conducted to verify the effectiveness and advantages of the proposed method. Full article

To address trajectory-tracking degradation of unmanned surface vehicles (USVs) in constrained waters caused by model uncertainty, strong environmental disturbances, and actuator limitations, this paper proposes a robust disturbance-observer-based optimization model predictive control method. First, a nonlinear tracking error model is established for a 3-DOF USV by incorporating environmental loads, parametric perturbations, and unmodeled dynamics into the kinematic and dynamic equations. Based on this model, a prediction model suitable for model predictive control is derived through linearization and discretization. Then, to estimate complex unknown disturbances online, a robust disturbance observer integrating a radial basis function neural network (RBFNN) with an adaptive sliding-mode mechanism is developed, enabling real-time approximation and compensation of lumped disturbances in the surge and yaw channels. Furthermore, to overcome actuator saturation caused by the direct superposition of feedforward compensation and feedback control in conventional composite strategies, a dynamic constraint reconstruction mechanism is introduced. By feeding the observer-generated compensation signal back into the MPC optimizer, the feasible control region is updated online so that the total control input satisfies both magnitude and rate constraints of the propulsion system. Theoretical analysis based on Lyapunov theory proves the uniform ultimate boundedness of the observation errors and neural-network weight estimation errors, while input-to-state stability theory is employed to establish closed-loop stability. Comparative simulations under sinusoidal trajectories, time-varying curvature paths, and large-maneuver turning conditions demonstrate that the proposed method significantly improves tracking accuracy, disturbance rejection capability, and control feasibility under severe disturbances and parameter mismatch. Full article

The craters on the moon hold important clues about the history of impacts in our solar system. To address the limitation of traditional intelligent methods in detecting buried craters, this study proposes a novel intelligent detection approach based on DEM and gravity data. We designed a hybrid network architecture (ResNet + ViT) that combines the local feature extraction strengths of Convolutional Neural Networks with the global context modeling capabilities of Vision Transformer. By combining the complementary information from DEM and gravity anomaly data, it achieves comprehensive detection of lunar craters—from those visible on the surface to buried subsurface structures. To mitigate the inherent sample imbalance in both gravity anomaly and DEM training data, we employ a U-Net architecture augmented with residual blocks and train it using a Focal Loss function with dynamic focusing parameters. Experimental results show that: (1) The proposed method attains high segmentation accuracy, achieving a mean Intersection over Union of 81.3% on the DEM test set and 82.6% on the gravity anomaly test set, respectively. (2) Our method outperforms U-Net and its mainstream variants, achieving a precision of 89.48% and superior detection completeness. (3) Application to representative geological units, including the Wugang Basin, Archimedes Crater, and Mare Moscoviense, validates the robustness and practical utility of our method. This study, thus, provides a novel technical framework for global-scale mapping of lunar impact craters and yields new insights into the evolutionary history of the lunar surface. Full article

This paper presents a unified analytical and computational framework for the study of typhoid fever transmission dynamics governed by a Caputo fractional-order compartmental model of order $\kappa \in (0,1]$. The population is stratified into five epidemiological classes, namely susceptible $\left(\mathcal{S}\right)$, asymptomatic $\left(\mathcal{A}\right)$, symptomatic $\left(\mathcal{I}\right)$, hospitalised $\left(\mathcal{H}\right)$, and recovered $\left(\mathcal{R}\right)$, and the governing system explicitly incorporates asymptomatic transmission, treatment dynamics, and temporary immunity with waning. The use of the Caputo fractional derivative is motivated by the well-documented existence of chronic asymptomatic Salmonella Typhi carriers, whose heavy-tailed sojourn times in the carrier state are naturally encoded by the Mittag–Leffler waiting-time distribution arising from the fractional operator. A complete qualitative analysis of the fractional system is carried out: the basic reproduction number ${\mathcal{R}}_0$ is derived via the next-generation matrix method; local and global asymptotic stability of both the disease-free equilibrium ${\mathcal{E}}_0$ (when ${\mathcal{R}}_0\le 1$) and the endemic equilibrium ${\mathcal{E}}^{*}$ (when ${\mathcal{R}}_0>1$) are established using fractional Lyapunov theory and the LaSalle invariance principle; and the normalised sensitivity indices of ${\mathcal{R}}_0$ are computed to identify transmission-amplifying and transmission-suppressing parameters. Existence, uniqueness, and Ulam–Hyers stability of solutions are established via Banach and Leray–Schauder fixed-point arguments. To complement the analytical results, a fractional physics-informed neural network ($\mathcal{PINN}$) framework is developed to simultaneously reconstruct compartmental trajectories and identify unknown biological parameters from sparse synthetic observations. $\mathcal{PINN}$ embeds the L1-Caputo discretisation directly into the training residuals and employs a four-stage Adam–L-BFGS optimisation strategy to recover five trainable parameters $\Theta$ = $\{\varphi ,\mu ,\sigma ,\psi ,\beta \}$ across three fractional orders $\kappa \in \{1.0,0.95,0.9\}$. The estimated parameters show strong agreement with the true values at the classical limit $\kappa =1.0$ ($\mathrm{MAPE}=2.27\%$), with the natural mortality rate $\mu$ recovered with $\mathrm{APE}\le 0.51\%$ and the transmission rate $\beta$ with $\mathrm{APE}\le 3.63\%$ across all fractional orders, confirming the structural identifiability of the model. Pairwise correlation analysis of the learned parameters establishes the absence of equifinality, validating that $\beta$ can be reliably included in the trainable set. Noise robustness experiments under Gaussian perturbations of $1\%$, $3\%$, and $5\%$ demonstrate graceful degradation ($\mathrm{MAPE}$: $0.82\%\to 3.10\%\to 7.31\%$), confirming the reliability of the proposed framework under realistic observational conditions. Full article

Spiking neural networks (SNNs) support energy-efficient machine intelligence because event-driven computation and sparse activity map naturally to low-power digital hardware. In practical implementations, however, membrane states, synaptic weights, and thresholds are represented with finite-precision integer arithmetic. Quantization, clipping, and overflow can therefore alter network dynamics rather than merely approximate a higher-precision model. This paper adopts an integer-state dynamical perspective, modeling a quantized SNN with a hardware-relevant update rule as a deterministic map on a bounded integer lattice. Rather than claiming recurrence itself as a new property, we focus on how finite-precision representation and implementation semantics shape observed recurrent regimes and activity patterns. We introduce a shift-based update rule with integer-valued states and investigate its behaviour through simulation-based analysis with network sizes $N=30$–130, connection densities 0.1–0.9, and bit widths 1 to 16 over T = 1000 steps. The results show bounded and recurrent temporal structure with strong quantization sensitivity. The observed regimes depend heavily on the semantics of representation and the scaling choices. These findings suggest that numerical precision can act as a dynamical design variable and provide useful implications for hardware-oriented SNN design, while motivating future work on attractor analysis and FPGA/ASIC validation. Full article

Transcranial magnetic stimulation (TMS) combined with electroencephalography (EEG) provides a powerful framework to probe and modulate human cortical and corticospinal excitability. In recent years, brain state-dependent EEG–TMS paradigms have gained increasing interest by synchronizing stimulation to ongoing neural activity. However, traditional approaches relying on single oscillatory features or fixed thresholds have yielded heterogeneous and often inconsistent results, motivating the adoption of machine learning (ML) and artificial intelligence (AI) methods to model brain state in a multivariate, data-driven manner. This review synthesizes current ML and deep learning (DL) approaches aimed at predicting cortical and corticospinal excitability from pre-stimulus EEG. We contextualize these methods within brain state-dependent EEG–TMS frameworks based on oscillatory phase, power, and network-level features, and within evolving definitions of brain state that move beyond local biomarkers toward distributed, large-scale, and dynamically evolving neural representations. The reviewed studies span feature-engineered models, data-driven decoding approaches, and emerging adaptive closed-loop frameworks. Finally, we discuss key methodological challenges, translational barriers, and future directions toward personalized, interpretable, and fully closed-loop neuromodulation systems. Full article

This study presents an optimized and comparative investigation of four intelligent energy management strategies—Proportional–Integral–Derivative (PID), Fuzzy Logic Control (FLC), Equivalent Consumption Minimization Strategy (ECMS), and Artificial Neural Network (ANN)—applied to a photovoltaic–fuel cell–battery hybrid electric vehicle (PV–FC–HEV). A high-fidelity MATLAB/Simulink model integrates a 6 kW proton-exchange membrane fuel cell (PEMFC), a 500 W photovoltaic subsystem with MPPT, and a lithium-ion battery (LiB) pack. While 1000 W/m² represents Standard Test Conditions (STC), the level of 400 W/m² was specifically selected to simulate average cloudy conditions common in urban driving environments, rather than standard NOCT (800 W/m²), to test the EMS’s robustness under significantly reduced PV support and stressed battery conditions (initial SOC = 30%). While surface contamination and the resulting performance degradation significantly impact real-world results, this study assumes a clean surface to establish an idealized performance baseline for the control algorithms. However, the authors acknowledge that contaminant accumulation is a key factor; future work will incorporate a degradation factor (e.g., a 10–15% efficiency penalty) to evaluate the reliability of these EMS strategies under actual operating conditions. ECMS achieved the lowest hydrogen consumption, saving up to 10 L compared with PID, while ANN maintained the most stable state of charge (SOC > 80%), minimizing deep discharge cycles and improving operational stability. FLC provided balanced operation under fluctuating irradiance. Overall, ANN offered the most harmonized energy flow and dynamic stability, whereas ECMS maximized fuel economy. The findings provide practical guidance for designing sustainable and intelligent control systems in next-generation hybrid electric vehicles. Full article