This paper investigates the adhesive bonding strength of steel-carbon fiber reinforced composites and their simulation analysis method, validating the approach on an automotive hybrid B-pillar to provide a basis for reliable joining of dissimilar materials. By combining adhesive bonding mechanics theory, the material property parameters of DC04 steel and carbon fiber reinforced polymer (CFRP) substrates are determined, offering parameters for finite element modeling of dissimilar material adhesive bonding. Lap shear and butt tensile tests are designed for the adhesive-bonded joints of DC04 steel and CFRP, enabling determination of main mechanical properties of Araldite 2015 structural adhesive and its adhesive performance. A finite element model of the DC04 steel-CFRP adhesive-bonded joint is built to simulate maximum load and failure displacement of the joint, analyze adhesive bonding strength, and compare results with test data. Scanning electron microscopy (SEM) is employed to analyze failure modes of the adhesive joints. A simulation analysis method for adhesive bonding of DC04 steel-CFRP dissimilar materials is established. Adhesive bonding simulations are performed on the automotive steel-CFRP hybrid B-pillar. Based on performance requirements of the B-pillar, simulations are conducted for axial tension, axial compression, lateral bending, and rearward bending, calculating deformation and stress parameters of the adhesive bonding elements under each condition. The carbon fiber composite B-pillar reinforcement plate is manufactured and bonded to the steel outer panel using Araldite 2015 adhesive. A three-point bending test was conducted on the hybrid material B-pillar assembly. Simulation values for displacement and stress are compared with test results, with maximum errors of 2.94% and 3%, respectively. The results validate accuracy and correctness of the adhesive bonding simulation for the dissimilar material B-pillar assembly and demonstrate feasibility and reliability of the proposed adhesive bonding simulation method and strength analysis for the steel and CFRP dissimilar material combination.

To solve the problems of low energy utilization efficiency, high equivalent hydrogen consumption, and significant power battery capacity degradation in hydrogen fuel cell tractors under row-crop seeding operations, a High-Efficient Low-Degradation Energy Management Control Strategy based on the Deep Deterministic Policy Gradient (HELD-DDPG) was proposed in this paper. Firstly, a tractor dynamics model was established based on the row-crop seeding operations. Subsequently, a power battery model was established based on the equivalent internal resistance theory to evaluate the state and energy consumption of the power battery, and a battery life model was established that accounts for the impact of capacity degradation on long-term system operation. A hydrogen fuel cell model was constructed according to the characteristics of the hydrogen fuel cell. To achieve a unified quantitative assessment of energy consumption across different energy sources in the entire power system, an equivalent hydrogen consumption model was established. On this basis, the HELD-DDPG was proposed with the objectives of minimizing the equivalent hydrogen consumption of the whole machine, stabilizing the SOC of the power battery, and minimizing battery capacity degradation, subject to the battery discharge current constraint. Finally, a rule-based energy management control strategy was selected as a comparative strategy. Simulation experiments on the HELD-DDPG were conducted using the MATLAB platform, and the effectiveness of the strategy was further validated through the hardware-in-the-loop (HIL) test platform. The simulation results showed that, compared with the comparative strategy, the HELD-DDPG significantly improved energy utilization efficiency of the tractor. Specifically, the chemical energy consumption of the hydrogen fuel cell was reduced by 45.43%, the cumulative energy throughput of the power battery was reduced by 32.92%, the total equivalent hydrogen consumption of the power system was reduced by 15.57%, and the battery capacity degradation was reduced by 19.76%. Meanwhile, the real-time implement ability of the HELD-DDPG was evaluated through the HIL test.

Traffic states evolve on irregular sensor graphs and vary with calendar context, yet the original ASTGCN does not explicitly model how the contribution of different graph receptive fields changes across traffic periods. This paper proposes CD-MRFG, a context-gated extension of ASTGCN that encodes hour-of-day, day-of-week and weekend information and uses the resulting representation to weight Chebyshev graph-convolution orders in each spatio-temporal block. Under a common 12-step forecasting protocol, CD-MRFG reduced the overall MAE and RMSE of the reproduced ASTGCN baseline from 18.66 and 31.05 to 16.98 and 28.59 on PEMS03, from 22.79 and 35.02 to 20.82 and 32.77 on PEMS04, and from 18.88 and 28.83 to 17.24 and 26.84 on PEMS08. Three-seed experiments confirmed lower mean MAEs on PEMS04 (p = 0.028) and PEMS08 (p = 0.042), although the corresponding RMSE differences did not reach the 0.05 significance threshold. Ablation, gate-weight, sensitivity, complexity and convergence analyses showed that temporal context was the main source of the improvement and that the gate provided a model-internal view of order selection with moderate overhead. CD-MRFG remains less accurate than several stronger recent baselines, so its value is a bounded and interpretable extension of ASTGCN rather than a universal state-of-the-art replacement.