Abstract:To address the issues in existing electric vehicle charging path planning methods, namely the neglect of queuing randomness and the difficulty in balancing real-time performance with global optimality, this paper proposes a path optimization method based on Generalized Total Cost (TGC) and a hierarchical Deep Q-Network (DQN). First, an M/M/c queuing model is introduced to quantify the randomness of charging waiting time. The TGC framework assists the Deep Q-Network in unifying nonlinear queuing time, travel time, and time-of-use electricity price into a single scalar reward. Second, a Top-M pre-screening mechanism combined with an action validity mask is employed to resolve convergence difficulties caused by large action spaces, and experience replay is utilized to break data correlations. The method""s performance is evaluated in a constructed high-fidelity simulation environment. When the Top-M parameter is set to 10, the proposed strategy achieves a global optimal solution hit rate of 89.5%, an average relative regret value converging to 0.0122, and a single inference time reduced to 0.35 ms.

Abstract:To investigate the effects of combined track gauge adjustment and rail grinding on the track state and vehicle dynamics of a mixed passenger and freight railway trunk line The state of rails, geometric parameters of tracks, and vehicle dynamics performances before and after track work in the experimental section were compared and analyzed. The results showed that the rail state was, and the personalized grinding eliminated the rail surface diseases, and no diseases occurred again within 6 months after grinding. The contact status between wheel and rail and the equivalent conicity were improved. The geometric parameters of the track were improved, and the gauge dynamic TQI decreased from 0.76 to 0.31 and 0.4 after the precise gauge adjustment and 3 months later, respectively. The gauge static TQI decreased from 0.7 to 0.15. The vehicle dynamics performances improved, and the maximum values of the transverse and vertical vibration acceleration of the EMU bogie decreased by 30.58% and 34.04% respectively. The derailment coefficient and the rate of wheel load reduction were significantly reduced, and the maximum values of the transverse and vertical vibration acceleration of the track inspection vehicle decreased by28.57% and 21.62%, respectively. Simultaneous implementation of rail gauge fine adjustment and rail grinding can significantly improve track quality and the dynamic performance of EMUs, ensuring the safety and ride comfort of train operation.

Abstract:During layered tunnel excavation, there is a complex three-dimensional spatial coupling relationship among bedding planes, tunnel contours, and stress directions, resulting in significant complex evolutionary characteristics of tunnel failure. Extrusion-based cementitious 3D printing technology offers advantages such as rock-like properties, rapid fabrication, low variability, and low cost. In this study, circular diversion tunnel models with different bedding orientations were fabricated using extrusion-based cementitious 3D printing. Overload failure tests and numerical simulations were conducted to analyze the mechanical characteristics and crack evolution mechanisms of layered circular tunnels under overload conditions. Results indicate that bedding orientation significantly influences the mechanical performance of layered tunnels. The peak strength of horizontal layered tunnels is higher than that of vertical ones. Bedding orientation primarily governs crack propagation paths. In horizontal bedding, cracks predominantly penetrate the layers, whereas in vertical bedding, cracks tend to propagate along weak bedding interfaces. For both orientations, crack initiation occurs at the tunnel crown and invert. Furthermore, horizontal layered tunnels exhibit a tensile-shear composite failure mode with matrix-penetrating cracks. In contrast, vertical layered tunnels show tensile failure along the bedding planes. These findings provide theoretical guidance for the excavation and support design of layered tunnels in stratified rock masses.

Abstract:【Objective】To address the imbalance in weight–volume resource utilization and the difficulty of coordinating vehicle loading and outsourcing decisions in multi-type vehicle transportation, a multi-type vehicle loading optimization method considering outsourcing is proposed.【Method】Cargo items are represented as two-dimensional weight–volume vectors, and a total-cost-minimization model is established. A genetic algorithm–dynamic volume–weight ratio decoding algorithm is proposed. The outer layer uses binary chromosomes to search for in-house and outsourced transportation decisions, while the inner layer constructs vehicle activation and cargo loading schemes according to vehicle operating costs and dynamic volume–weight ratios.【Result】For random instances with = 10, 20, and 30, the algorithm yields solutions with relative deviations of no more than 0.84% from the Gurobi optimal solutions, and the result for = 10 coincides with the Gurobi optimal solution. In the enterprise case, allowing outsourcing reduces the total cost by 14.67% and decreases the number of activated vehicles from 13 to 5.【Conclusion】The proposed algorithm coordinates in-house loading and outsourcing decisions, improves vehicle resource allocation efficiency, and supports transportation cost reduction and operational decision-making.

Abstract:To address the problems of small scale, blurred boundaries, weak features and complex background interference in insulator flashover trace detection from UAV inspection images, an improved YOLOv8-based method is proposed. CBAM and a receptive field enhancement structure are introduced into the Backbone to strengthen feature representation of key trace regions. A weighted bidirectional feature fusion structure with a preserved P2 branch is constructed in the Neck to enhance the fusion of shallow details and deep semantic features. A P2 small-object detection head and detail enhancement blocks before the P2/P3 branches are added in the Head to improve localization and recognition of tiny flashover traces. A hard-sample closed-loop optimization strategy is further used to improve model stability. Experimental results show that the proposed method can effectively detect insulator flashover traces and has potential for engineering application.

Abstract:To address the technical bottlenecks of natural Goos-Hänchen (GH) shift, including small amplitude, weak signal intensity, and difficulties in detection and practical application, this study leverages the ultra-high Q-factor and strong light field localization characteristics of bound states in the continuum (BIC) and quasi-BIC. With silica as the substrate and hafnium dioxide rectangular nanorods as the core structure, a novel all-dielectric reflective BIC metasurface is proposed by embedding a hafnium dioxide thin film between the substrate and nanorods. Combined with numerical simulation and theoretical analysis, the Q-factor is calculated by fitting the reflection spectrum with the Fano formula, and the modulation laws of structural parameters and incident angle on resonance characteristics, reflection phase and Goos-Hänchen shift are systematically investigated. The results show that quasi-BIC with a high-quality factor can significantly enhance the GH shift, with the maximum GH shift reaching more than 400 times the resonant wavelength. The maximum GH shift exactly occurs at the reflection peak with unit reflectivity, which facilitates the detection of GH shift signals, and the corresponding refractive index sensing sensitivity reaches 3.58×106μm/RIU. This research realizes the giant enhancement and multi-dimensional modulation of the Goos-Hänchen shift, providing a new scheme for the practical application of micro-nano photonic devices such as ultra-sensitive refractive index sensors.

Abstract:In the daily safety inspection of railway infrastructure based on Unmanned Aerial Vehicles (UAVs), the perception of encroaching obstacles, such as gravel, is crucial for ensuring operational safety. However, the highly irregular edge morphology of such obstacles, coupled with the strict computational resource constraints in practical engineering deployment, makes it difficult for existing models to balance feature extraction efficiency with shape-adaptive perception. This paper proposes a customized YOLO11 detection architecture trained entirely from scratch (with zero pre-trained weights). In the backbone network, a Lightweight Efficient Pyramid Adapter (LEPA) is designed to decouple features and significantly strip away redundant parameters. During the multi-scale fusion stage, a Deformable Attentional Multi-scale Fusion (DAMF) module is introduced, which combines spatial attention masks with deformable offsets to adaptively and precisely conform to the physical contours of irregular targets. Experimental validation demonstrates that, compared to the baseline YOLO11, the proposed method boosts the mAP@0.5 and mAP@[0.5:0.95] by 1.72 and 5.12 percentage points, respectively. Benefiting from feature decoupling, the parameter count and theoretical computational complexity are significantly reduced by 5.41% and 11.11%, respectively. This study effectively overcomes the detection challenges caused by limited computational resources and variable target shapes, providing a lightweight and highly efficient technical solution for high-precision intelligent perception in complex railway environments.

Abstract:This study 7075 aluminum alloy was used as the substrate. Ceramic coatings are fabricated via micro-arc oxidation (MAO) under regulated frequencies of 100, 400, 700 and 1000 Hz, followed by hydrophobic modification with stearic acid. The effects and mechanisms of frequency on microstructure, surface wettability and corrosion resistance of coatings are systematically investigated. The results reveal that MAO frequency greatly alters the micromorphology and porosity of coatings. The coating prepared at 400 Hz possesses the densest structure with the minimum porosity of approximately 1.24%. By contrast, the porosity rises to around 2.13% and 3.95% at 100 Hz and 1000 Hz respectively. After stearic acid modification, the surface contact angle increases remarkably from 43.2°—59.1° to 117.8°—123.4°, endowing the coatings with low surface energy. The modified coating obtained at 400 Hz achieves the maximum contact angle of 123.4° and the lowest surface energy of 15.45 mJ/m2. Electrochemical tests demonstrate that the MAO coating formed at 400 Hz exhibits the lowest corrosion current density of (1.05×10-9 A/cm2) and the highest polarization resistance of (4.91×107 Ω·cm2). Excellent corrosion resistance is maintained after hydrophobic modification, with the polarization resistance reaching about (1.45×107 Ω·cm2). Neutral salt spray tests show that no obvious corrosion occurs on modified coatings after 504 hours, while corrosion products accumulate on unmodified specimens. It is confirmed that MAO frequency adjusts discharge behavior to tailor pore structure and surface state, thereby determining hydrophobic modification efficiency and corrosion protection capacity. The coating prepared at 400 Hz delivers optimal comprehensive properties. This work provides a theoretical reference for the design and fabrication of high-performance protective coatings on aluminum alloy surfaces.

Abstract:To address the challenges of source number estimation and the susceptibility of weak sources to masking during multi-point concurrent fault acoustic source localization in traction power supply systems, this paper proposes a novel MUSIC-based localization method employing a spiral array and adaptive eigenspectrum entropy estimation. The proposed method formulates an entropy model based on the global distribution of the eigenspectrum and introduces an eigenvalue squaring enhancement mechanism, which facilitates the precise determination of concurrent source numbers under complex background noise conditions. Simulations replicating colored noise and mechanical vibration environments typical of traction substations demonstrate that the proposed algorithm significantly outperforms the traditional Minimum Description Length (MDL) criterion in both estimation accuracy and convergence speed, particularly under conditions of low signal-to-noise ratio (SNR) and limited snapshots. Furthermore, in wide-dynamic-range scenarios where the weak source constitutes merely 20% of the total signal power, the method maintains robust and accurate source estimation. Extensive Monte Carlo simulations corroborate that this approach substantially enhances the spatial resolution and localization accuracy for multiple concurrent fault acoustic sources, thereby exhibiting considerable potential for practical engineering applications.

Abstract:Under the context of smart capacity expansion, dynamically opening the emergency lane has been gradually adopted to alleviate heavy traffic pressure on expressways, showing relatively stable performance under normal conditions. However, once a traffic accident occurs, the emergency lane no longer serves solely as a traffic space but must also accommodate emergency response vehicles, and its control strategy directly affects traffic safety and the subsequent recovery process.In existing practices, the lack of differentiated responses to various accident scenarios often makes it difficult to balance traffic operations and emergency access. To address this issue, this study starts from the impact of accidents and quantifies accident severity by introducing an accident severity index. A fuzzy clustering method is then applied to classify accident levels. Based on the classification results and the location of the accident, corresponding emergency lane control strategies are developed. For high-risk segments such as bridges, stricter control measures are adopted to ensure operational safety.For validation, multiple accident scenarios are constructed on the SUMO platform, and three strategies—lane closure, permanent opening, and graded dynamic control—are compared. The results indicate that, while ensuring the passage of emergency vehicles, the graded dynamic control strategy can improve traffic operations under incident conditions, with more stable performance observed in high-risk scenarios. These findings provide a reference for the practical management of emergency lanes under smart capacity expansion conditions.