The floating charge condition of lithium-ion batteries widely exists in scenarios such as backup power sources and communication base stations, and is a special condition that tends to stabilize. This stability poses a challenge to the quantitative diagnosis of internal short circuit (ISC) in batteries under this condition. In this study, a quantitative diagnosis method for early ISC in lithium-ion battery packs based on intermittent charging was proposed. This method utilized a repeated “charging-rest” process to calculate the equivalent leakage current according to the relationship between charging capacity and leakage, thereby achieving rapid quantitative diagnosis of ISC. The simulation and experimental results show that the proposed method has a diagnostic accuracy of less than 2% and a detection time of about 33 minutes for a micro-ISC battery with an ISC resistance of 500 Ω. It achieves early-stage and high-precision quantitative diagnosis of battery ISC under floating charge conditions. In addition, compared with conventional constant voltage source methods, the proposed method improves the accuracy of diagnosing short circuits within 100 Ω by at least 16 times. The proposed ISC method has a very low computational burden and is of great significance for improving the safety of battery packs.

Efficient geographic and geological knowledge services for railway engineering form a crucial foundation for supporting the multi-scale and multi-disciplinary intelligent applications of digital twin technology in railways. To improve the completeness of query results and enhance the capability for integrated analysis across the multi-scale “region–engineering project–construction site” applications of railway digital twins, a meta-network model for the dynamic distribution of geographic and geological knowledge was proposed. This model was designed as a distributed knowledge base network system, with railway agents, business departments, knowledge relationships, and multi-scale application scenarios as key nodes. A meta-network disruption algorithm for distribution optimization was implemented, with node importance assessed using degree centrality indicators. By analyzing the distribution influence through network perturbation, the influence range of nodes was calculated, obtaining the optimized distribution structure of the knowledge base. To validate this approach, it was applied to optimize the distribution of the knowledge base within the knowledge base management and application scenario for railway digital twins of a major railway bridge project. Experimental results show that, when processing knowledge retrieval tasks at the engineering and regional scales, the distribution optimization method increases the number of query results with reduced retrieval time and enhanced result matching accuracy.

The pile-net composite foundation can effectively reduce the post-construction settlement in road engineering, and has often been used for ground improvement for runway and apron systems in the airfield of coastal airports in China recently. However, the available research on the pile-net composite foundation is mostly focused on the bearing characteristics before construction, lacking research to quantitatively evaluate the influence of aircraft taxiing loads on the bearing characteristics of the pile-net composite foundation in the runway. Therefore, the parameters such as the thickness of the fill layer, the spacing between piles, and the type of aircraft were changed. In addition, based on the dynamic load influence coefficient, the differential settlement ratio after construction, and the degradation coefficient of soil arch, the finite element software ABAQUS was used, and a numerical analysis of the mechanical deformation characteristics of the pile-net composite foundation in the runway under the influence of aircraft taxiing load was conducted. The influence of aircraft taxiing load on the bearing and settlement characteristics of the pile-net composite foundation in the runway was quantified. The results show that when the thickness of the fill layer decreases from 4.5 m to 1.5 m, the soil arching effect is weakened by about 3.3% to 15.1%. When the spacing between piles decreases from 6d to 3d, the soil arching effect is weakened by about 7.8% to 12.0%. The differential settlement ratio after construction at the location of the soft soil layer may exceed the recommended value of the code. As the fill layer becomes thinner, the pile spacing and aircraft weight are greater; the main gear load is more concentrated, and the soil arching effect becomes weaker.

To clarify the relationship between rockfall impact force and impact load in civil engineering, a comprehensive reflection coefficient is introduced to reflect the interaction between falling rocks, cushioning soil layers, and structures (a higher value indicates a more significant impact effect on the structure). A rockfall impact load calculation method based on wave theory is proposed. The size and distribution characteristics of rockfall impact loads were studied through experiments on arched structures with overlying cushioning soil layers subjected to rockfall impacts. The values and influencing patterns of the comprehensive reflection coefficient were obtained. The proposed calculation method was used to analyze the relationship between rockfall impact loads and rockfall impact forces. Through the research, it was clarified that the rockfall impact load on the structure presents a symmetrical parabolic distribution on the cross-section and can be characterized by a quadratic parabolic curve equation controlled by the maximum impact pressure peak at the arch crown and the structural span. The results indicate that within a range of 10 meters of free fall height for falling rocks and a cushioning soil layer thickness of 2.0 meters, there is a significant negative correlation between the comprehensive reflection coefficient and the thickness of the cushioning soil layer. When the thickness of the cushioning soil layer is 2.0 meters, its influence on the shape and free fall height of the falling rock is relatively small, and a value of 0.55 can be used. When the thickness is 1.0 meter and 0.5 meters, the comprehensive reflection coefficient of cuboid falling rocks is greater than that of spherical or conical falling rocks, and it is positively correlated with the falling height of the rocks. The relationship between the resultant force of rockfall impact loads on the structure and the impact force of rocks on the cushioning soil layer depends on the thickness of the cushioning soil layer and the shape of the falling rocks. When the thickness of the cushioning soil layer is 2.0 meters, the two are close, with the former slightly smaller than the latter, making the resultant force of rockfall impact loads equal to the rockfall impact force on the structure, which leans towards safety in structural design. However, when the thickness of the cushioning soil layer is less than 2.0 meters, the reverse is true, and the smaller the thickness, the greater the difference. When the thickness of the cushioning soil layer is 1.0 meter, the average increase of cuboid falling rocks is about 20 times larger than that of spherical or conical ones, and when the thickness is 0.5 meters, the average increase is about 30 and 10 times respectively for the two shapes. Under the same conditions, the resultant impact load of cuboid falling rocks is greater than that of spherical or conical ones, and the difference between them increases as the falling height of the rocks increases or the thickness of the cushioning soil layer decreases.

Rapidly obtaining co-seismic landslide distribution and conducting disaster assessments after earthquakes are vital for effective emergency relief and reconstruction. Therefore, in this study, the InSAR data-Newmark physical fusion driver model (IDNPM) was used to rapidly assess the landslides triggered by the earthquake in Jishishan, Gansu Province on December 18, 2023, with a view to quickly and accurately grasping the macroscopic distribution of landslide hazards. Firstly, through the time series satellite-based augmentation system (SBAS)-InSAR, it was revealed that there was serious gully development and retrogressive erosion in this area. These geological characteristics provided a favorable breeding environment for landslides. Secondly, the IDNPM was used to quickly evaluate the landslide of Jishishan earthquake, and it was predicted that the steep slopes and gully sides of Zhaomuchuan Village, Tashapo Village, and Dahejia Town were the high-risk areas for earthquake-induced landslide. Finally, based on the field investigation, numerical simulation, and satellite identification technology, the reliability of the model in practical application was proven. The results indicate that a total of 2.657% of the region is at high risk. There is a need to focus on such zones by urgently clearing and stabilizing slopes where landslides have occurred. For areas where no landslides have occurred, monitoring and assessment measures should be taken to guard against possible post-earthquake secondary landslide events. The research results can provide strong data support for emergency relief and reconstruction work after earthquakes in the affected areas.

To address the damage and deterioration of concrete materials in a sulfate environment in marine engineering, a concrete erosion test and microscopic test under different concentrations of sulfate solution and high temperature in the semi-immersed environment were carried out. By analyzing the mass change, dynamic elastic modulus, composition and content of eroded products, and sulfate ion content of concrete specimens and conducting the microscopic test, the damage and deterioration law of concrete under the high temperature and dry−wet cycle in semi-immersed and sulfate environment was revealed. The results show that in the early stage of sulfate erosion, sulfate invades the interior of concrete in the immersion zone and promotes hydration. The ettringite (AFT) and gypsum inside the concrete fill the initial gap, and the products play a filling and skeleton role. However, the continuous increase of the product amount in the later stage causes damage to the concrete, and the dynamic elastic modulus decreases by 12%–30%. The concrete in the adsorption zone is subjected to both the dry−wet cycle and the thermal cycle. The two cycles aggravate the capillary and diffusion effects so that the salt is easy to crystallize inside the concrete. Additionally, the thermal cycle accelerates the separation of aggregate and cement matrix, forming more pores, which is conducive to the continued intrusion of salt. The filling effect in the early stage of sodium sulfate crystallization improves the strength, and the mass increases by 0.5%–1.75%. However, the damage effect in the later stage is more serious.

To improve the cohesiveness of foam asphalt cold recycled binder (composed of cement, old asphalt, and foam asphalt), three modifiers of foam asphalt cold recycled binder, including polyphosphoric acid (PPA), polymerized styrene butadiene rubber (SBR), and triglyceride (TG) were selected, and molecular models of foam asphalt cold recycled binder with different modifiers and their dosage were established by molecular dynamics simulation method. The effect of different modifiers and their dosage on the improvement of the interface properties of foam asphalt cold recycled binder was studied based on interface energy, interaction energy, diffusion coefficient, and radial distribution function (RDF) of water molecules. The significance difference of each index and the weight of the modification scheme were discussed by significance tests and the analytic hierarchy process. The results have shown that the addition of 0.8% PPA has the most significant effect on the cohesiveness of the foam asphalt cold recycled binder, which can effectively improve the interfacial energy, free volume, and water damage of the foam asphalt cold recycled binder, and also make the foam asphalt cold recycled binder have better diffusion ability. The addition of 15.0% TG can improve the interaction effect of the foam asphalt cold recycled binder, but the diffusion ability is poor. The addition of 2.0% SBR has the best effect on the water damage resistance of foam asphalt cold recycled binder, but the overall modification effect of SBR is not so good. There is no significant influence among different dosages of modifiers, and each modification scheme is independent and has its own advantages. After comprehensive comparisons, 0.8% PPA has the best effect on improving the interfacial cohesiveness of foam asphalt cold recycled binder.

In order to study the mechanical properties of aeolian sand concrete (ASC) under axial compression with different basalt fiber (BF) contents, the experiment took concrete with 0% fiber content as the reference group and designed the basalt fiber–aeolian sand concrete (BF-ASC) with different fiber content of 0.05%, 0.10%, 0.15%, and 0.20%. The influence of the BF volume fraction on the axial compressive strength, the peak stress, the peak strain, and the elastic modulus of ASC was analyzed. The results show that the peak stress and elastic modulus of the concrete first increase and then decrease with the fiber volume fraction. At 0.10% BF, the peak stress and elastic modulus of the concrete are 115.6% and 112% of the reference group, respectively. The peak strain and toughness indices increase with the volume fraction of the fiber. At 0.20% BF, the peak strain and toughness indices increase by 34.92% and 7.2% compared to the reference group. The entire stress–strain curve of BF-ASC undergoes three stages of elastic, elastoplastic, and failure, just like ordinary ASC. The BF-ASC constitutive relation can be described in terms of segments based on the Carreira and Chu model and the Guo Zhenhai’s model. The ascending profile is consistent with the Carreira and Chu models, and the descending profile is consistent with the Guo model. In addition, a regression analysis is conducted on the BF-ASC uniaxial compressive constitutive model considering the BF content. The correlation coefficients are all greater than 0.98, and the model has a high degree of agreement with the experimental curves.

Ultra-high ductile concrete (UHDC) has excellent strain hardening and multi-cracking characteristics, and it has great potential in impact load resistance. Direct tensile tests were conducted under 11 strain rates (0.000 1–189.0700 s⁻¹) ranging from quasi-static to impact states to investigate the strain rate effect on the tensile properties of UHDC. The influence of strain rate on the shape of the tensile stress–strain curves, cracking pattern, and tensile performance indicators of UHDC was analyzed. The expression of the dynamic increase factor of tensile performance indicators regarding strain rate was established. In addition, the influence of the tensile rate on the fiber–matrix interface bonding performance was analyzed to further explain the strain rate effect on the tensile properties of UHDC. The results show that the deformation capacity of UHDC decreases with the increase in strain rate. When the strain rate is 102.000 0 s⁻¹, UHDC still has significant strain hardening and multi-cracking pattern capacity, with the tensile strain capacity up to 4%. The average crack width, which keeps a constant value of 100 μm, does not vary with the strain rate change, exhibiting the excellent crack control capacity of UHDC. The relationship curves between the dynamic increase factor of the tensile performance indicators and strain rate show a clear two-stage characteristic.

Self-sensing concrete materials for health monitoring have emerged as a new research focus in the field of structural engineering, yet there are challenges in the progress of their application and industrialization. To promote the application of self-sensing concrete in structural health monitoring, research on the effects of various conductive fillers on the performance of the concrete from the aspects of the dosage ratio, shape characteristics, secondary modification, and mixing with other kinds of fillers was introduced, and the significant achievements and milestones in the development of functional fillers for self-sensing concrete were reviewed. The testing and calibration standards for self-sensing functional fillers are not well-established. Different testing equipment and methods can significantly influence detection results, making it challenging to ensure comparability of results. Environmental adaptability assessments are inadequate. Complex environmental conditions (temperature, humidity, corrosion, etc.) have a substantial impact on material durability and service life, and there is insufficient research on the long-term stability of materials in actual operation. Quality control during mass production has not received sufficient attention. Disparities in raw materials and processes in large-scale production can severely affect the consistency of product performance. There are limited real-world engineering application cases. Conducting operational trials of intelligent self-sensing concrete with real-time multi-parameter monitoring and multifunctional coupling in large structures such as bridges and tunnels can supplement relevant data, offering promising research prospects for self-sensing concrete.

In order to reduce the impact of vehicle-induced vibration on the surrounding environment of high-speed railway bridges, firstly, a spherical bearing with built-in metal disc isolators for vertical vibration isolation of bridges was proposed, and the mechanical constitutive model and overall vertical stiffness calculation method of the bearings were given. The design compliance of the vertical stiffness of the metal disc isolator was studied by combining simulation and experimental comparison. Simulation was also used to analyze the influence of different friction support surfaces on the vertical stiffness and stress of the metal disc isolator. Secondly, through a series of tests on spherical bearings for vertical vibration isolation, the conventional performance, vertical stiffness, isolation load, dynamic and static stiffness ratio, and stiffness stability of the bearings were studied. Finally, taking the 32-meter-span concrete simply supported box girder of high-speed railway as the background, the vibration isolation effect of the bearings under the action of trains was explored. The results show that the deviation between the simulated and measured vertical stiffness values of the metal disc isolator with three different vertical bearing capacity designs is within ±10%. The friction coefficient of the bottom support surface of the metal disc isolator ranges from 0.01 to 0.10, with a 2.1% increase in vertical stiffness and a 0.9% decrease in stress. The conventional performance of the spherical bearing for vertical vibration isolation meets the design requirements, and the deviation between the test values and the design values of vertical stiffness and isolation load is less than ±10%. The vertical stiffness of the bearing remains stable after overload and unloading, providing protection for the metal disc isolator. At an excitation frequency of 1–17 Hz, the range of the dynamic to static stiffness ratio of the bearings is 1.00–1.15. The stiffness increase of the bearings after 10 million fatigue cycles is less than 10%, and all components are intact. The bridge supported by spherical bearings for vertical vibration isolation meets the requirements of train safety and riding comfort indicators. The soil vibration response attenuation of the former bridge is about 4 dB higher than that of the bridge supported by ordinary spherical bearings.

In order to investigate the influence of the number of ballast particle fracture surfaces on the direct shear performance of the ballast bed, pebbles were crushed to obtain ballast aggregates with different numbers of fracture surfaces. Direct shear tests were carried out under different vertical stresses to obtain the stress–strain relationship and deformation characteristics of ballast aggregates. Based on formula calculation, the variation laws of peak shear strength and internal friction angle of ballast aggregates with vertical stress were obtained. The results show that, in terms of strength characteristics, when the number of fracture surfaces is 0, 1, 2, and 3, the shear strength and internal friction angle of ballast aggregates under the same vertical stress increase with the number of fracture surfaces. The shear strength increases by 14.7%–25.6%, 12.2%–27.4%, and 6.0%–10.1%, respectively. In terms of deformation characteristics, the shear shrinkage of ballast aggregates increases with fracture surface number, with a maximum increase of 76.4%, while the shear dilation decreases with a maximum decrease of 20.8%.

To accurately and efficiently evaluate the vibration damping performance of tracks, the German standard DIN V 45673-4 provides a method for efficiently calculating the insertion loss rate, which can eliminate interference from stochastic factors such as line operation conditions and track irregularities. However, this method neglects the contribution of track bending stiffness in its equivalent model, leading to errors in identifying the resonant frequency of the wheel-rail system and the vibration damping frequency band of tracks. To increase the calculation precision, the model for calculating the German-standard insertion loss rate was improved on the premise that the computational efficiency was maintained. By employing the elastic foundation beam theory to approximate the actual structural dimensions and stiffness characteristics of tracks and considering the contributions of elastic components and track bending stiffness, the stiffness of the model was corrected. The plans for optimizing track dimensions and vibration damping performance of elastic components were discussed in terms of the 1–80 Hz vibration damping frequency band required by China’s environmental impact assessment by taking the floating slab tracks with vibration damping pads for example. The results show that after stiffness correction, the error in the resonant frequency of the wheel-rail system is reduced from 76.0% to 4.9%. The increase in the fastener loss factor can achieve the overall vibration damping performance within the 1–80 Hz frequency band. However, within the range of 30–90 kN/mm for fastener stiffness, reducing (increasing) the fastener stiffness improves the insertion loss rate of the 60–80 Hz (30–60 Hz) frequency band and decreases the rate of the 30–60 Hz (60–80 Hz) frequency band. This does not meet the overall vibration damping performance for the 1–80 Hz frequency band which can only be achieved when the fastener stiffness exceeds 90 kN/mm. Furthermore, lower stiffness of vibration damping pads or thicker floating slabs indicate higher improvement in fastener stiffness.

To address the limitations of difficult convergence to the global optimal solution with fixed step search, strong dependence on the selection of initial growth point, and huge growth space for the plant growth simulation algorithm (PGSA), a new strategy for search mechanism of adaptive variable step, Gauss perturbation mutation mechanism, and screening mechanism of growth space was proposed. On this basis, the PGSA based on multi-mechanism fusion (multi-mechanism fusion PGSA) was established. The prestress optimization of suspended domes was further carried out by using the multi-mechanism fusion PGSA and compared with other algorithms. The results show that compared with the original PGSA, the introduction of search mechanism of adaptive variable step can avoid the algorithm falling into local optimal solutions; the introduction of Gauss perturbation mutation mechanism can solve the problem of poor optimization results caused by the improper selection of initial growth points, and the introduction of screening mechanism of growth space can effectively terminate the growth after the algorithm converges, thus significantly reducing the growth space by 97.64%. The number of iterations of multi-mechanism fusion PGSA is the smallest (only 45), and the absolute value of the average horizontal radial reaction of supports after optimization is minimal (only 0.004 kN) in comparison with other algorithms. Therefore, the applicability of this algorithm is verified.

To address the problems of local buckling of cold-formed thin-walled steel members and cracking of ceramsite lightweight concrete, while considering the aesthetic requirements for interior building layouts, a prefabricated cross-shaped column partially encased with cold-formed thin-walled steel and filled with lightweight concrete was proposed. To investigate the seismic performance of these special-shaped columns, four cross-shaped columns were designed and fabricated using different coarse aggregate replacement rates as a parameter, and the low cyclic reversed loading tests were carried out. Based on the experimental study, the finite element software ABAQUS was used to analyze the lightweight aggregate concrete strength, steel plate strength, steel plate thickness, and loading angle. The test results indicate that the hysteretic curves of the four specimens are symmetrical and full, exhibiting a shuttle shape. A compression-bending failure mode of the specimens was observed. As the coarse aggregate replacement rate increases from 0% to 30%, 70%, and 100%, the specimen weight decreases by 77 kg/m³, 176 kg/m³, and 252 kg/m³, respectively; the carbon emission reductions decrease by 19.18%, 38.11%, and 49.93%; the ultimate load-bearing capacity decreases by 1.0%, 4.7%, and 9.2%; the ductility coefficient increases by 1.4% at first, and then decreases by 3.8% and 4.2%; the energy dissipation reduces by 8.6%, 2.5%, and 6.7%. The use of fly ash ceramsite to replace ordinary stone as a concrete coarse aggregate has no significant effect on the seismic performance of the columns but shows great potential for carbon emission reduction compared to ordinary concrete. Increasing the strength of lightweight aggregate concrete does not significantly improve the load-bearing capacity, ductility, or energy dissipation performance of the specimens. When the steel plate strength increases from Q235 to Q355, the ultimate load-bearing capacity of the specimens increases by 45.1%. When the steel plate thickness increases from 4 mm to 5 mm and 6 mm, the ultimate load-bearing capacity of the specimens increases by 14.8% and 35.5%, respectively. The least favorable loading angle for the specimens is 45°.

Forest fire detection is crucial for forest fire emergency rescue. To address the shortcomings of existing models in sample quality, multi-scale issues, and generalization capability across multi-view images, a method for forest fire detection based on YOLO (FFD-YOLO) was proposed. First, a multi-view visible light image dataset for detecting forest fire from of high point view (FFHPV) was constructed to enhance the model’s learning capability for multi-view fire information. Second, omni-dimensional dynamic convolution was introduced to develop an omni-dimensional spatial pyramid pooling (OD-SPP) to improve the model’s feature extraction capacity for multi-view fire characteristics. Finally, a wise intersection over union (Wise-IoU) loss function with a dynamic non-monotonic focusing mechanism was introduced to mitigate the impact of low-quality data on model precision and enhance small-target fire detection. Experimental results have demonstrated that FFD-YOLO increased precision by 3.9%, recall by 3.7%, mean average precision (mAP) by 4.0%, and F₁-score by 0.038 compared to YOLOv7. In comparative experiments with YOLOv5, YOLOv8, dense distinct query (DDQ), detection transformer with improved denoising anchor boxes (DINO), Faster R-CNN, Sparse R-CNN, Mask R-CNN, FCOS, and YOLOX, FFD-YOLO attained 75.3% precision, 73.8% recall, 77.6% mAP, and 0.745 F₁-score, validating its feasibility and effectiveness.

To address the issues of repeated decoding and resource waste caused by random batch generation of the outer code in existing outer code block coding schemes for batched sparse (BATS) codes, the optimization of theoretical batch count and dynamic adaptability of BATS codes was systematically investigated based on the outer code block encoding scheme. First, under the condition of a known packet loss rate, a batch consumption analysis model for BATS codes was established, and an optimal degree value computation method was derived to tackle the challenges in existing schemes regarding theoretical batch count calculation and optimal degree value determination for minimizing batch count consumption. Second, for scenarios with unknown packet loss rates in the channel, a reinforcement learning-based dynamic degree optimization method for BATS codes was proposed, enabling real-time acquisition of degree values through an intelligent learning mechanism. Finally, simulation experiments were conducted to evaluate the theoretical model and the proposed dynamic optimization method. Simulation results have shown that the established transmission model based on outer code blocks and its batch count computation formula can be used to calculate batch consumption and determine the optimal degree distribution. Simulation results demonstrate that the proposed reinforcement learning-based optimization scheme achieves lower average batch count consumption than fixed-degree value schemes with unknown packet loss rates and maintains great performance in dynamic channel environments.

To meet the needs of human-robot collaboration, where an inspection manipulator actively cooperates with a person under the railroad car and to enhance the convergence speed of the proximal policy optimization (PPO) algorithm, an adaptive PPO (a-PPO) algorithm was proposed and innovatively applied in the online motion planning of the inspection manipulator. Firstly, the system model was designed to immediately output policy actions based on the current environmental state. Secondly, geometric reinforcement learning was introduced to construct the reward function, utilizing the agent’s exploration to continuously optimize the distribution of rewards. Thirdly, the clipping value was adaptively determined based on the policy similarity between before and after the update, and the a-PPO algorithm was developed. Finally, the improvement effects of the a-PPO algorithm were compared on two-dimensional maps, and the feasibility and effectiveness of its application were experimentally verified in both simulation and real train scenarios. The results indicate that in the two-dimensional plane simulation, the a-PPO algorithm shows certain advantages in convergence speed compared to other PPO algorithms. Additionally, the stability of paths has been improved, with the average length standard deviation being 16.786% lower than that of the PPO algorithm and 66.179% lower than that of the Informed-RRT* algorithm. In the application experiments in both simulated and real train scenarios, the manipulator demonstrates the capability to dynamically adjust target points and actively avoid dynamic obstacles during motion, reflecting its adaptability to dynamic environments.

To address the power frequency overvoltage and high costs caused by cable applications in high-voltage dedicated line continuous power supply systems, an optimized hybrid line configuration scheme was proposed. First, a no-load equivalent circuit model was established using two-port network theory, and the distribution laws of voltage and no-load circulating currents along the hybrid lines were theoretically derived by incorporating the distributed parameters of the lines. Second, a π-type equivalent circuit of the system was constructed through port equivalence, and a unified multi-load power flow model was developed using a network splitting algorithm. A calculation method for system equivalent impedance and the traction network’s maximum supply distance was proposed to quantitatively evaluate the power supply capability of hybrid lines. Furthermore, to minimize the total life cycle cost, the proportion of cable and overhead line installations was optimized. Simulation results demonstrate that the “cable + overhead line” scheme effectively suppresses power frequency overvoltage, reduces no-load current to one-third of that in pure cable configurations, and retains the long-distance supply advantage of cables (maximum supply distance of 95 km), saving approximately 5.1 million yuan in investment compared to conventional schemes.

To resolve the difficulty in controlling heavy-haul trains operating in complex environments caused by insufficient driver experience, a multi-mass dynamic model for multi-locomotive traction was established based on the Locotrol synchronous control principle of the Datong–Qinhuangdao Railway. A controller was designed for the main locomotive, where the total time-varying unknowns, including coupler forces, running resistance, and external disturbances, were regarded as aggregated uncertainties. The acceleration of these uncertainties was further treated as an extended state, enabling real-time estimation and compensation via an extended state observer. Moreover, the fast terminal sliding mode control was introduced to improve the nonlinear error feedback control law in active disturbance rejection control, and an improved adaptive reaching law was employed to refine the dynamic quality of the sliding mode reaching motion. Simulations were conducted on a heavy-haul train with the formation of “1 + 105 + 1 + 105 + controllable end” by incorporating actual line data from Datong–Qinhuangdao Railway and expert driver experience, and compared with traditional methods. The simulation results demonstrate that, compared to conventional sliding mode active disturbance rejection control, the proposed method reduces control force chattering in master-slave locomotives by 23.7%, improves tracking accuracy by 19%, and confines tracking errors within ±0.7 km/h.

The dynamic load distribution and stiffness characteristics of bearings are important factors that cause vibration and cutting stability of machine tools. On the basis of nonlinear elastic Hertz contact theory and Jones-Harris model, combined with the judgment criterion of roller–raceway contact state, a five-degree-of-freedom analysis model of multi-row combined ball bearings was proposed to compare dynamic load distribution and dynamic stiffness characteristics of left and right rows in tandem back-to-back (TBT) combined ball bearings under different operating conditions. The improved iterative algorithm was adopted for solution, which greatly improved the convergence of the iterative calculation under the fluctuation of different external conditions and yielded the dynamic load distribution and dynamic stiffness characteristics within the TBT combined bearings under constant preload. The results show that the rotational speed, radial load, and axial load can change the load distribution and dynamic stiffness characteristics of combined bearings, and have different effects on the left-row and right-row bearings. The increase in rotational speed can lead to the incomplete contact area within the bearings, which makes the load distribution of the bearing oscillate. Compared with that of a single bearing under equivalent external load conditions, the radial dynamic stiffness of the combined bearing is more significantly improved than the axial dynamic stiffness.

To optimize the individualized comprehensive performance of plug-in hybrid electric vehicles (PHEVs), an optimization method for the shift schedule of single-shaft parallel PHEVs considering both dynamic and economic performance while reflecting the driving intention was proposed. Firstly, the switching logic among different operating modes was determined according to the demand torque, the engine characteristic curves, and the state of charge (SOC) of the power battery, and torque distribution strategies under different operating modes were formulated. Subsequently, a fuzzy inference method was used to establish a quantitative model for driving intention, which could calculate the driver’s expectations for dynamic and economic performance based on the driver’s operation and vehicle status. Then, by taking the driver’s expectations for dynamic and economic performance as the weights of corresponding sub-objective functions, a linear weighting method was used to construct a comprehensive performance evaluation function, thereby optimizing the shift schedules under different driving intentions. Finally, a simulation model was developed using MATLAB/Simulink, and simulations under the WLTC test cycle were conducted with initial SOC values of 0.5 and 0.9, respectively, using the optimal dynamic, optimal economic, and individualized optimal shift schedules. Simulation results show that under both SOC initial conditions, while reflecting the driving intention, the equivalent fuel consumption (EFC) of the individualized optimal shift schedules is reduced significantly compared to that of the optimal dynamic shift schedule, with reductions of 10.1% at an SOC of 0.5 and 11.8% at an SOC of 0.9. Meanwhile, the EFC of the individualized optimal shift schedules is increased compared to that of the optimal economic shift schedule, with increases of 5.3% at an SOC of 0.5 and 1.7% at an SOC of 0.9.

The middle and rear axles of three-axle mining dump trucks are the main load-bearing axles, and their output characteristics influence the driving stability of the vehicles. To solve the load-bearing problem of mining dump trucks and improve the driving stability of the vehicles at the same time, a kind of hydraulically interconnected suspension was proposed. The impedance transfer matrix method was adopted to derive the flow-pressure relationship of the suspension, and the 1/2 vehicle mechanical-hydraulic coupling equation was obtained. The complete decoupling of body-wheel motion modes was realized by solving the system state matrix. Finally, the road obstacle experiment was carried out to investigate the operational characteristics of the suspension. The analysis results show that the suspension system reduces the natural frequency of body bounce motion, increases the damping ratio, attenuates the body bounce vibration quickly, and improves the dynamic load distribution of the middle and rear wheels. When the vehicle passes a 120 mm high triangular road obstacle at the speed of 15 km/h, the maximum pressure difference between the two oil cylinders of the hydraulically interconnected suspension is 1.15 MPa, and the maximum response time is 0.3 s. These are good to achieve uniform load. At the same time, the displacement of the two oil cylinders is basically equal, but the direction is opposite, which plays a good role of displacement compensation and helps to maintain the attitude of the body during travel.

In order to effectively characterize the mechanical response of vehicle-end contact and coupler instability during high-speed train collisions, a contact force calculation method for the longitudinal impact of a thin-walled structure was established, which considered the impact of vehicle misalignment on contact force and contact moment. Characterization methods for the load characteristics of different coupler-buffer devices after the crushing stroke were then established. Finally, based on these two methods, a three-dimensional (3d) train collision dynamics model was constructed considering the vehicle-end contact and coupler instability. The dynamics model results were compared with those from the finite element model (FEM) under two working conditions. The results show that the proposed method can better reflect the contact force responses of different thin-wall structures at different impact speeds, with a maximum relative error of 9.83% compared to the finite element results. The characterization method can effectively distinguish the post-crushing load characteristics of different types of coupler-buffer devices. The developed 3D collision dynamics model is in good agreement with the finite element calculation results in key indicators such as vehicle speed responses, collision interface force responses, and vertical responses of the carbody.

To further investigate the synergistic layout of aerodynamic braking devices on high-speed trains operating at 400 km/h and above and to clarify the overall braking benefit and efficiency of aerodynamic braking systems suitable for China’s current standard trainsets, the shape of the CR400AF platform and the configuration of its basic braking system were taken as a reference. Different numbers of “butterfly-type” aerodynamic braking devices were installed, and the aerodynamic characteristics of high-speed trains equipped with such devices under different operating conditions were simulated and calculated. A direct integration method applicable to aerodynamic braking problems was proposed. The braking performance of trains relying solely on aerodynamic braking devices at a given initial speed was compared with coasting to stop, and the braking effect was analyzed. Train braking equations were established, and the segmented accumulation method was used to calculate the braking distance and time when aerodynamic braking was combined with service braking and emergency braking. The results show that the installation of aerodynamic braking devices significantly increases the overall aerodynamic resistance of trains, and higher layout density leads to stronger aerodynamic interference between front and rear braking plates. The composite braking method combining aerodynamic braking effectively compensates for insufficient adhesive braking force at high speed, while addressing the low braking efficiency of aerodynamic braking at low speed. The combined braking distance is proportional to the square of the speed, and the braking time is proportional to the speed. With combined aerodynamic braking, the emergency braking distance from an initial speed of 350 km/h can be reduced to 5500 m or below.

To make route guidance strategies more targeted for enhancing efficiency and practability, a route guidance model based on a multi-layer network and vehicle-source information of the regional highway was proposed. Firstly, a multi-layer network of the regional highway was constructed based on the complex network theory. The congested segments in the highway network were identified, and their vehicle sources were located and clustered. The locations for distributing route guidance information were further determined. Next, the travel cost function was controlled by applying the social welfare coefficient. A route guidance model with a variable parameter was established for balancing individual and social benefits. Finally, a route guidance information release framework was constructed, and the influence of the proportion of travelers using the guided routes on the system was investigated when the route guidance scheme was implemented. The results have shown that the proposed model guides a few travelers, and their average travel time increases by 2.1 minutes, while the average travel time of all travelers decreases by 9.1 minutes. The generated route guidance scheme poses a small adverse impact on the travelers, effectively decreases the total time spent when considering the fairness, and provides a more efficient and feasible strategy for alleviating highway congestion.

To reduce the phenomenon in which China-Europe Railway Express operators compete for cargo sources by offering excessively low prices, an asymmetric tripartite evolutionary game model of “government-operator-supplier” was constructed under the reward-and-punishment mechanism of the government. The model was based on local government subsidies to China Railway Express operators and suppliers and took into account the actual conditions in the operation process of China Railway Express. Through numerical simulation, the influence of the main factors, including the reward and punishment of the government, on the evolutionary stability strategy of the system was analyzed, providing theoretical references for the coordinated development of the China Railway Express accordingly. The results have shown that the increased government punishment will effectively promote synergistic development between operators and suppliers. Differential subsidies are given according to the different strategies of operators and suppliers. As the difference between different subsidy levels becomes larger, it is more beneficial for synergistic development. Reasonable setting of the maximum subsidy per unit of container provided by the government to operators, and gradually reducing the amount until the subsidy is completely withdrawn, is an effective way to promote synergistic development. The additional social benefits of the government are a positive influence in determining the choice of government strategies.

The interaction between natural disasters forms a complex disaster chain, making the losses caused by composite disasters more severe. To quantify the risk of disasters caused by complex disaster chains, explore the vulnerability level of regions to complex disaster chains, and effectively promote disaster risk prevention work, the triggering and superposition (reduction) effects of disaster chains were considered. A vulnerability assessment index system for composite disaster systems was constructed from three dimensions: exposure of disaster-bearing bodies, susceptibility, and adaptability of disaster-prone environments. A series of models for assessing the exposure degree of composite disaster-bearing bodies, disaster-prone environmental sensitivity, and adaptability was established through derivation. Subsequently, they were weighted to obtain a series of vulnerability assessment models for composite disaster systems. By taking the rainstorm-landslide disaster chain in the Guangdong−Hong Kong−Macao Greater Bay Area as an example, the vulnerability index of rainstorm, landslide, and rainstorm-landslide disaster chain in the Greater Bay Area was calculated by combining convolutional neural network （CNN）, coupling model of a parameter optimal geographical detector and analytic hierarchy process （OPGD-AHP）, G1 method-Technique for Order Preference by Similarity to Ideal Solution (G1-TOPSIS), and entropy weight-TOPSIS, and the corresponding vulnerability level zoning map was further drawn by using ArcGIS tools. The research results indicate that the vulnerability of the rainstorm-landslide disaster chain is high and relatively high in the western region, medium in the central and western regions, as well as southwest and northeast regions, and low and relatively low in the central, central and southern regions, and eastern regions in the Greater Bay Area. There are not only overlapping relationships but also certain triggering and synergistic effects between the vulnerability of different single disaster types in the same region, especially between high vulnerability areas, as well as between low vulnerability areas. The results can be promoted and applied in the vulnerability assessment of composite disaster systems, providing technical support for risk assessment and disaster reduction and prevention of composite disaster systems in China.