武汉城市职业学院

Single Image Super-Resolution (SISR) aims to recover high-resolution details from low-resolution inputs, a task that remains challenging due to the significant loss of high-frequency information. To address limitations in cross-dimensional feature interaction and detail restoration, this paper proposes a Multi-scale Hybrid Attention Residual Network (MHARN). MHARN integrates three novel modules: a Progressive Convolution Group (PCG) that employs parallel dilated branches for dynamic multi-scale shallow feature aggregation; a Multi-head Hybrid Attention Module (MHAM) that combines channel, spatial, and positional information to capture detailed multi-dimensional features and long-range dependencies; and a Dynamic Feature Enhancement Module (DFEM) that generates parameterized kernels for adaptive feature extraction based on local image content. We evaluate the proposed model on five standard benchmark datasets: Set5, Set14, BSD100, Urban100, and Manga109. Extensive quantitative and qualitative experiments demonstrate that MHARN achieves excellent Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) performance. The results show that MHARN significantly reduces visual artifacts and recovers sharper textures compared to state-of-the-art methods, providing a robust solution for high-fidelity image reconstruction.

Efficient thermal management is crucial for maintaining the durability and electrochemical stability of proton exchange membrane fuel cells (PEMFCs). A three-dimensional (3D) multiphysics model was developed to investigate the coupled heat-fluid-electrochemical behavior of PEMFCs using a micro-encapsulated phase change material suspension (MePCS) as coolant. The effects of micro-encapsulated phase change material (MePCM) volume fraction (phi v), coolant velocity (Vcoolant), and inlet temperature (Tin) on thermal and electrochemical performance were systematically analyzed. Results show that incorporating MePCS effectively enhances temperature uniformity and suppresses local overheating. Compared with pure water, the maximum membrane temperature difference (Delta Tmax) and temperature uniformity index (TUI) decreased by 68.2% and 42.3%, respectively. The optimal operating parameters were identified as phi v= 15%, Vcoolant= 0.05 m/s, and Tin = 340.15 K, achieving a net power density of 0.518 W & sdot;cm-2. Moreover, the co-flow configuration of the reaction gas and coolant improved power density by 1.41% over counter-flow, attributed to better heat-water-gas coupling. These findings demonstrate that MePCS based coolants provide an effective approach to improving PEMFCs water and thermal uniformity and overall efficiency.

The diffusion of information in social networks is jointly shaped by network topology and the interaction among multiple types of information. Structurally, social networks contain not only pairwise relationships but also numerous group-based interactions. Dynamically, different information types may facilitate or inhibit each other during diffusion. However, most existing models rely on low-order networks built from pairwise links, limiting their ability to capture higher-order effects arising from group interactions. To address this limitation, this study proposes a novel information diffusion model that integrates simplicial complexes and multilayer coupling mechanisms, enabling the coexistence of individual-and group-level interactions as well as the mutual influence among different information types. The model extends the classical unaware-aware-unaware (UAU) paradigm and considers four node states to represent the interactive diffusion of two types of information. Using the Microscopic Markov Chain Approach (MMCA), we derive a set of dynamic evolution equations and analyze the critical diffusion threshold via mean-field theory. Simulation results verify the accuracy of the theoretical analysis and demonstrate that higher-order structures obviously reduce the threshold and enhance diffusion performance. Moreover, the model reveals how mutual facilitation or inhibition between information types and the density of higher-order structures jointly affect the scale and speed of diffusion. This work provides new insights into the co-evolutionary dynamics of information and has implications for designing effective information propagation strategies in real-world complex networks.

Aiming at the problems of long test cycle of classical transfer path analysis method and low test accuracy of operating condition-based transfer path analysis method, as well as the unique NVH challenges faced by electric vehicles (EVs) such as unmasked structure-borne road noise and complex vibration transfer paths derived from powertrain electrification, a method of interior structure-borne road noise diagnosis which can take into account both test efficiency and test accuracy. Based on the tested noise transfer function (NTF) and the acceleration admittance data at the car body attachment point, the acceleration-sound pressure transfer relationship of each path of the car body was obtained, and combined with the acceleration frequency response at the car body attachment point under operating conditions, the sound pressure of the vehicle structure-borne road noise was synthesized, and the total sound pressure was evaluated. The contribution of each path’s sound pressure response to the total sound pressure response is studied, and the main transmission path of 56 Hz peak noise is identified by combining NTF and modal analysis method. The feasibility and effectiveness of the method are verified by experiments: first, the main transmission path of peak noise is accurately predicted through the proposed sound pressure synthesis method; then, targeted modification is carried out based on the prediction result; finally, the peak noise is reduced by 4.19 dB(A).

Corner-module electric vehicles equipped with four-wheel independent drive, four-wheel independent steering, and electromechanical braking (EMB) actuators provide a flexible platform for software-defined chassis control, but braking-in-turn maneuvers impose severe longitudinal–lateral coupling and competition for tire adhesion resources. This paper proposes an actuator-oriented hierarchical coordinated control strategy for EMB-based corner-module vehicles. At the upper level, a Model Predictive Controller optimizes lateral tire force allocation under a tire-friction-ellipse hard constraint and coordinates the four-wheel steering response. At the lower level, a three-intensity adaptive braking-force distribution algorithm converts the vehicle-level demand into wheel-level EMB clamping-force commands while considering braking intensity, steering intensity, load transfer, and yaw stability. To improve actuator tracking accuracy, the EMB subsystem combines nonlinear actuator modeling, offline parameter identification, online recursive-least-squares correction, and force–speed–position cascade control. MATLAB (R2025b)/Simulink-CarSim co-simulation and EMB hardware-in-the-loop (HIL) tests verify the proposed strategy under fixed-angle emergency braking and lane-change braking conditions with high, low, and variable-adhesion roads. The results show improved trajectory tracking and yaw stability, reduced braking-torque fluctuation, and faster EMB clamping-force response, demonstrating the suitability of the proposed actuator-level coordination method for intelligent electric chassis applications.