Search Results (655)

Hairpin windings are increasingly adopted in permanent magnet (PM) traction machines owing to their high slot fill factor, compact end-winding structure, and suitability for automated manufacturing. However, limited heat dissipation and high copper losses under peak loading and high-frequency operation result in severe thermal constraints, which restrict the power rating of the machine. This paper presents a multi-physics comparison of different winding cooling topologies for a PM machine with hairpin winding, including hollow conductor cooling, end-winding cooling, and cooling channel insertion at slot-bottom, slot-middle, and slot-opening regions. A coupled electromagnetic–thermal model based on the finite element method (FEM), which accounts the heat transfer between different components, is used to analyze temperature distribution, losses, efficiency, loading capacity, and hydraulic requirements. The results show that the position of the cooling channel has great influence on the thermal behavior and electromagnetic performance of the machine under different working conditions. The study emphasizes the strong coupling between cooling design, conductor geometry, AC loss behavior, and efficiency and provides practical design guidelines for selecting appropriate cooling techniques in high-power-density traction machines. Consequently, an improved cooling system results in a reduced amount of PM for the same output power range. Full article

Spatial models used in architecture, civil engineering, and geodesy require topological consistency to support downstream applications such as energy simulation, indoor navigation, and urban planning. Existing Boolean operation engines process individual mesh objects but do not guarantee the consistency of a complete space partition, particularly when unbounded space must be represented explicitly. This paper presents a general embedding algorithm that inserts an arbitrary polyhedral boundary representation B into an existing space partition A, which constitutes a manifold, gapless partition of Euclidean space including unbounded cells, through a sequence of topologically controlled split operations. Unbounded edges are treated as rays and intersected directly by the geometric kernel, so no separate virtualization of the geometry at infinity is required. The algorithm proceeds in three phases: face-by-face comparison with tolerance-based classification to detect all intersections, incremental collection of split data for edges, faces, and cells through a multi-pass traversal, and execution of split work steps on the topological kernel in ascending dimension order. By operating exclusively through split operations, validity is checked incrementally at each split rather than by a separate global repair pass, so A is kept a valid space partition throughout without any smoothing or repair of an invalid result. A systematic test case matrix covering nine categories of volumetric relationships and boundary contact types was developed. Selected representative scenarios from this matrix were implemented and checked for topological correctness on the resulting partition, including preservation of manifold connectedness and gap-free space coverage without post hoc repair. The algorithm generalizes the parametric design operators introduced in a preceding conference paper to the full three-dimensional case and provides a theoretical foundation for consistent spatial modeling of the built environment. Full article

Axial-flux permanent-magnet (AFPM) machines are becoming an increasingly promising solution for electromechanical systems requiring high power density. In particular, their use is expanding to electric vehicles (EVs), the aerospace industry, and advanced industrial applications, such as renewable energy applications. Their compact design, high torque-to-mass ratio, and relatively high efficiency make AFPM machines an attractive alternative to traditional radial-flux solutions. However, their integration for widespread application remains limited due to challenges in design, manufacturing, thermal management, and control systems, which ultimately also have an economic impact. This article presents a comprehensive and systematic review of AFPM machines, covering key aspects, including topology classification, design methodologies, electromagnetic modelling, optimisation methods, materials and manufacturing processes, and advanced control strategies. A structured, multi-level classification of AFPM machines is presented, incorporating stator and rotor configurations, magnetic circuit structures, winding types, and materials, thereby providing a unified overview of existing designs. Furthermore, the article presents an in-depth analysis of the sizing equations used to calculate and estimate the parameters, approaches to electromagnetic modelling (including the finite element method and magnetic equivalent circuits), and modern optimisation methods based on artificial intelligence. Particular attention is paid to materials science and new manufacturing technologies, such as soft magnetic composites, printed circuit board stators, and additive manufacturing, as well as to thermal management solutions required for high-power-density applications. This work provides a unified reference framework for researchers and engineers and outlines future directions for the development and industrial adoption of AFPM machines. Full article

This paper investigates a near-field millimeter-wave (mmWave) integrated sensing and communication (ISAC) system under unknown interference. A base station equipped with a partially connected dynamic subarray hybrid architecture serves a legitimate user while performing target-oriented transmit beampattern shaping. Unlike existing works that assume perfect interference knowledge, we characterize the unknown interference channels via a robust spatial covariance uncertainty model. To exploit spatial degrees of freedom for interference suppression, the user employs a fully connected hybrid receiver. We formulate a robust transmit power minimization problem subject to worst-case communication signal-to-interference-plus-noise ratio (SINR) and sensing beampattern constraints, alongside constant-modulus and dynamic subarray hardware constraints. To solve this highly non-convex mixed discrete–continuous problem, we propose a two-layer alternating optimization framework. The inner layer optimizes the continuous and phase-quantized beamformers using successive convex approximation, while the outer layer refines the binary subarray connections via a penalty-augmented local discrete search. Extensive simulations demonstrate that explicitly modeling worst-case uncertainties ensures reliable ISAC performance in adversarial environments, and the dynamic subarray architecture systematically outperforms conventional fixed topologies in power efficiency. Additional robustness and sensitivity analyses show that these gains are most pronounced when sufficient spatial degrees of freedom remain, whereas excessive antenna failures, unmodeled strong multipath, or covariance drift outside the uncertainty envelope can erode the communication and sensing margins. Full article

Topological descriptors play an important role in chemical graph theory and QSPR/QSAR studies by relating molecular structure to measurable physicochemical properties. Among standard benchmark families, octane isomers are frequently used to evaluate the behavior of degree-based descriptors because of their rich branching patterns and well-documented physicochemical data. Although many studies have examined topological indices for octane isomers, comparatively fewer works have focused on topological coindices and derived coindex-based descriptors. In this work, we study several known topological coindices and four derived descriptors, denoted by $K{J}_1,K{J}_2,K{J}_3,$ and $K{J}_4$, for comparative analysis on the octane-isomer benchmark. We also present a unified treatment of lower and upper bounds for the first and second $(\alpha ,\beta )$-general Zagreb indices, together with their reduced and expanded variants, in terms of basic graph parameters, such as the minimum degree, maximum degree, order, and size. These bounds cover a range of familiar special cases, including classical Zagreb, forgotten, Sombor, and Randić-type indices, thereby placing several known descriptors within a common framework. For the application part, the original octane-isomer analysis is retained as a controlled benchmark for the proposed descriptors. In addition, an expanded QSPR experiment is added using a Zenodo molecular dataset containing 90 organic compounds and nine physicochemical endpoints. SMILES strings were converted into hydrogen-suppressed molecular graphs, graph-theoretical descriptors were computed, and ordinary least squares, ridge regression, and PLS(2) models were evaluated using an 80:20 train/test split and five-fold cross-validation. The expanded results show strong or useful performance for selected endpoints, especially critical volume, molecular volume, standard Gibbs free energy of formation, and logarithmic water solubility, whereas some temperature-related endpoints remain less stable. The results therefore support the usefulness of degree-based and coindex-based descriptors as compact exploratory QSPR variables while also emphasizing the need for cautious interpretation, redundancy analysis, and external validation on broader chemical families. Full article

Decarbonizing heavy-duty logistics requires powertrains that integrate novel topology design, degradation-aware optimization, and robust dynamic performance under real-world operational loads. While solid oxide fuel cells offer high efficiency, their application in transportation is hindered by thermal fatigue. This study proposes a novel hybrid powertrain topology integrating a metal-supported solid oxide fuel cell (SOFC), a micro gas turbine (MGT), and an aluminum–silicon phase change material (PCM) thermal buffer. A high-fidelity dynamic model is developed and coupled with a multi-objective optimization framework to size the PCM buffer and battery pack, balancing capital expenditure and system lifetime. Furthermore, a degradation-aware energy management strategy based on a thermal state-of-charge metric is introduced. Simulations over a 10 h dynamic drive cycle indicate that the optimal configuration (120 kg PCM, 80 kWh battery) extends the SOFC’s simulated remaining useful life to 38,400 h, a 2.5-fold improvement over unbuffered systems. Concurrently, the proposed energy management strategy reduces the MGT mechanical wear index by 98% compared to conventional load-following strategies. The system demonstrates robust performance across ambient temperatures from −20 °C to +45 °C and achieves a 22% reduction in projected capital expenditure compared to standard proton exchange membrane fuel cell powertrains. This topology offers a highly durable and economically viable pathway for next-generation zero-emission heavy-duty vehicles. This work addresses a critical gap in the literature: the lack of integrated thermal buffering and degradation-aware control strategies for high-temperature fuel cell systems in dynamic vehicular applications. By coupling a physical latent heat buffer with a novel Thermal-SOC-proportional Energy Management Strategy, the proposed architecture directly targets the primary degradation mechanisms that have historically impeded SOFC commercialization in heavy-duty transport. Full article

Low-frequency sound insulation remains a major challenge for conventional passive materials, as improved attenuation is usually achieved at the expense of increased thickness and mass. In this work, a smooth fixed third-order ring-curve fractal-maze acoustic metamaterial is proposed for compact low-frequency sound insulation, and a physics-guided long short-term memory–physics prediction surrogate network (LSTM–PPS-Net) tandem framework is developed for its full-spectrum inverse design. Different from conventional Hilbert-type, right-angled, or sharply folded labyrinthine structures, the proposed topology uses recursively arranged curved channels to extend the effective acoustic propagation path and enhance phase accumulation within a limited space. Based on this mechanism, four physically meaningful parameters, namely slit width d, characteristic radius R₃, wall thickness t_w, and inter-column spacing l_E, are selected to construct a low-dimensional design space. A COMSOL–MATLAB automated finite-element method (FEM) workflow is established to generate 1000 valid transmission-loss (TL) spectra over 100–1700 Hz with a 5 Hz interval. For forward prediction, PPS-Net is developed by integrating geometry encoding, frequency-conditioned spectral decoding, and peak-weighted learning. The proposed PPS-Net achieves the best prediction accuracy among the tested models, with a mean absolute error (MAE) of 0.75 dB, a root mean square error (RMSE) of 1.88 dB, and a coefficient of determination (R²) of 0.96, outperforming multi-layer perceptron (MLP), convolutional neural network (CNN) and Transformer models under the same dataset and training protocol. For inverse design, the LSTM encoder extracts frequency-ordered spectral features from the target TL curve, while the frozen PPS-Net decoder provides differentiable acoustic-response feedback, thereby addressing the non-unique mapping from acoustic response to structural parameters. Furthermore, a compactness-oriented optimization strategy is introduced to balance spectral consistency, peak alignment, bandwidth preservation, and occupied-area reduction. In two representative cases, the optimized designs reduce the occupied area by approximately 21% in both representative cases, while maintaining the target attenuation characteristics after FEM verification. These results demonstrate that the proposed framework provides an efficient and physically interpretable route for the full-spectrum inverse design and compact optimization of low-frequency acoustic metamaterials. Full article

Item-to-group recommendation identifies the most compatible user groups for a specific item provider to enable precision marketing, such as recommending fruit products to the most receptive consumer communities. Existing graph-based recommendations typically treat social relationships as static binary links, failing to capture variations in interaction intensity driven by user preferences. Moreover, these models largely overlook the structural relevance of intra-group connections, leading to unreliable group representations. To address these challenges, we propose the Preference-Weighted Neighbor-Aware Group Recommendation Network (PNGRN). Specifically, social edges are first reweighted using preference signals derived from historical user–item rating interactions, thereby suppressing socially connected but preference-inconsistent neighbors during aggregation. Structurally cohesive candidate groups are then identified via k-core decomposition, retaining only subgraphs where every member has at least k internal connections. A neighbor-aware graph convolutional network (GCN) module is further introduced to incorporate external social neighborhood features into group representations. This ensures that the learned group profiles reflect both internal structural stability and the external social context. Experiments on three real-world datasets demonstrate that PNGRN consistently outperforms competitive baselines across all evaluation metrics. Notably, on the MovieLens-1M dataset, PNGRN achieves up to a 9.85% improvement in Precision@20 and a 8.98% gain in NDCG@20. These results validate the necessity of coupling topological density with external social influence, and this work offer a scalable framework for precision group-targeted marketing. Full article

Degree-based topological indices play a central role in characterizing graph structures and their chemical applications. Among these, multiplicative Zagreb indices have attracted considerable attention due to their strong discriminative power and relevance in chemical graph theory. Neighborhood versions of these indices extend the classical concept by incorporating the aggregate degree information of adjacent vertices, capturing more subtle structural effects related to local branching. Trees, as connected acyclic graphs, provide a natural and tractable class for studying the extremal behaviors of these indices, while molecular trees—trees with a maximum degree of at most four—serve as chemically meaningful models of acyclic organic compounds. Investigating extremal values on these structures offers both theoretical insight into the indices’ behavior and identification of molecular graphs that maximize or minimize them. In this work, we determine the maximal and minimal values of the neighborhood-based multiplicative Zagreb indices for trees of fixed order and prescribed maximum degree, and we provide a complete structural characterization of all extremal graphs. Special attention is given to molecular trees, for which explicit extremal bounds are derived and all optimal structures are identified. These results provide efficient tools for evaluating the indices and illuminate the structural principles governing their extremal behavior. Full article

Magnetic skyrmions have emerged as promising candidates for next-generation nanomagnetic devices owing to their stability, nanoscale size, and efficient manipulability. In this work, we demonstrate the deterministic creation of skyrmions using a single ultrafast laser pulse in a thin ferromagnetic film. Through micromagnetic simulations, we model the effect of a focused picosecond laser pulse on a Pt/Co-based multilayer with interfacial Dzyaloshinskii–Moriya interaction (DMI). We find that above a threshold laser fluence, or equivalently, a critical pulse duration, a stable 25 nm Néel-type skyrmion diameter is created at low temperature under a modest out-of-plane magnetic field. Our results demonstrate that skyrmions can be written deterministically by a single picosecond laser pulse, eliminating the need for multiple exposures or electrical stimuli. This work systematically identifies the ultrafast excitation and material-parameter ranges that enable stable solitary skyrmion nucleation in experimentally realistic magnetic multilayers. This can be a foundation for photonic-spintronic integration, enabling optical data writing and magnetic storage, offering a pathway toward ultrafast, energy-efficient, and contactless control of topological spin states for future memory and logic applications. Full article

Due to the influence of the control strategy of the battery energy storage station (BESS), the degree of voltage sag, and the battery state of charge (SOC), the fault current characteristics of the BESS outlet line differ significantly. Traditional overcurrent protection faces the risk of failure to operate. To evaluate the operational performance of overcurrent protection of outlet line connecting BESS, this work first analyzes the topological structure and control strategy of BESS and further investigates the fault current characteristics of its outlet line. Based on this, the operational performance of overcurrent protection relay is studied. In addition, an overcurrent protection method of outlet line connecting BESS considering the battery SOC is proposed. By calculating and setting the SOC boundary, reliable protection of outlet line within different SOC intervals is achieved. Finally, a grid-connected model of BESS is built based on an electromagnetic transient simulation software to verify the operational characteristics of the proposed method. Full article

In this paper, we present an applied numerical study inspired by recent theoretical advances on boundary feedback control for fractional coupled PDE–ODE systems. While earlier works have mainly focused on proving the existence, uniqueness, and stability of solutions within the fractional Lyapunov framework, our contribution lies in translating these theoretical results into a practical setting for graph neural networks (GNNs). In this model, the partial differential equation describes the diffusion of information signals across the network topology, while the fractional-order ordinary differential equation captures the nonlinear and memory-dependent update rules of the node states. By employing the backstepping algorithm, we design a boundary controller that guarantees Mittag–Leffler stability of the coupled system. Numerical simulations demonstrate that, in the absence of control, the network dynamics exhibit instability and divergence, whereas the proposed boundary control gradually stabilizes the information propagation process. These results underline the effectiveness of the method and its potential relevance for the fractional modeling and regulation of graph-based neural architectures. An example is given with a consensus problem that shows that the boundary controller stabilizes the network dynamics. Full article

Molecular property prediction is a fundamental task in drug discovery and materials design. While graph neural networks (GNNs) and SMILES-based Transformers have made significant strides, the former are often limited by local message-passing bottlenecks such as over-squashing, while the latter frequently lack explicit topological constraints and suffer from severe vocabulary imbalance. In this work, we revisit the granularity of molecular modeling and propose a representation learning framework built upon bond-level sequences. Our framework models molecules as sequences of directed bond tokens and introduces a structure-aware hybrid attention mechanism. By imposing hard topological constraints on a subset of attention heads to reinforce local connectivity while preserving global receptive fields in the remaining heads, the design is intended to separate short-range chemical bonding from long-range contextual dependencies. For pre-training, we implemented a multi-scale consistency learning paradigm, which utilizes an atom-centric group masking strategy to induce a hierarchical loss of local structural information and employs contrastive and triplet losses to ensure identity consistency across varying scales of structural degradation. Furthermore, by incorporating macro-scale physicochemical descriptors (e.g., LogP, TPSA) as global anchors, we examined how the inclusion of global attribute bias can provide weak physicochemical priors during pre-training, while its effect during downstream fine-tuning remains task-dependent. Experimental results demonstrate that our lightweight model, with approximately 3.5 million parameters, exhibits a dataset-dependent performance profile across MoleculeNet benchmarks and shows promising behavior on selected topology-sensitive tasks, particularly MUV. Ablation studies further analyze the contribution of bond-level connectivity, the stage-dependent dynamics of global attribute bias, structured masking, and pre-training configurations. Ultimately, this work provides an alternative representation design for molecular modeling, offering a parameter-efficient option for future molecular learning systems alongside traditional SMILES-based and graph-based formulations. Full article

Traditional urban building energy modeling often overlooks the complexity of spatial configurations and mutual shading effects, thereby limiting its accuracy. This study proposes a novel, interpretable, data-driven framework based on heterogeneous graph neural networks (GNNs) to uncover and characterize the complex interrelationships between building morphology and urban topology. Using a parametric platform, this study generated a graph dataset of 285 residential blocks in Wuhan, structured as a dual-level graph: Building Zone Graphs (BZGs) and Building Layout Graphs (BLGs). Four GNN models were trained based on the dataset, and the evaluated results demonstrate that GraphTransformer outperforms GCN, GAT, and GraphSAGE in capturing long-range spatial relationships―particularly those arising from shading and solar access interactions. On a validation set, GraphTransformer achieved superior predictive accuracy, with R² scores exceeding 0.85 and 0.90 for cooling and heating energy predictions, respectively. After that, post hoc interpretability analysis by GNNExplainer identified three important morphology features influencing building energy consumption. Critically, the model found that shading relationships encoded as graph edges―especially those between southern and western façades―had statistically significant influence on building energy consumption. Finally, this work establishes an efficient, interpretable surrogate modeling framework for urban-scale energy analysis, delivering quantifiable, design-actionable insights to support sustainable urban development. Full article

Accurate short-term traffic flow prediction is crucial for managing macroscopic Intelligent Transportation Systems (ITS). To overcome limitations in capturing complex spatiotemporal dependencies and the severe challenges of hyperparameter tuning, this paper proposes IHO-CNN-BiLSTM-Attention, a novel hybrid deep learning framework. Specifically, a Convolutional Neural Network (CNN) extracts local spatial features, a Bidirectional Long Short-Term Memory (BiLSTM) network captures temporal dependencies, and an attention mechanism dynamically weights key timesteps. To maximize the architecture’s performance, an Improved Hippopotamus Optimization (IHO) algorithm is proposed for automatic hyperparameter optimization. The IHO algorithm effectively overcomes the premature convergence of traditional optimizers by integrating a Piecewise Linear Chaotic Map (PWLCM) for initialization, tangent-based non-linear adaptive weights, a Tangent Flight defense mechanism, and Lens Opposition-Based Learning (LOBL) for local optimum escape. Evaluated comprehensively across three distinct macroscopic traffic benchmark datasets (a multimodal intersection, METR-LA velocity, and PeMSD4 volume), the IHO algorithm first demonstrated statistically significant superiority on standard CEC benchmark functions. Subsequently, the proposed hybrid model achieved state-of-the-art traffic state classification performance, maintaining peak F1-Scores of 0.9798, 0.8436, and 0.9561 across the highly diverse datasets. It significantly outperformed both classical optimized baselines (e.g., PSO, GWO) and contemporary heavy deep learning architectures (e.g., ASTformer, DiffSTG) under severe class imbalance and varying topological conditions. This work offers a robust, scalable, and highly generalized spatiotemporal forecasting solution with strong theoretical guarantees for intelligent traffic control. Full article