Abstract:This study presents a joint identification method for structural parameters， input loads， and system states based on the Gaussian process latent force model （GPLFM）. The structural parameter identification is integrated into the hyperparameter optimization process of GPLFM， thereby avoiding additional model updating steps， and the load identification accuracy significantly outperforms that of mainstream Bayesian filtering methods. In the proposed method， the unknown load time history is first modeled as a Gaussian process. By converting the Gaussian process regression into a linear state-space model， the numerical model， measurement data， and prior information of input loads are organically integrated. A Bayesian filter is then employed to estimate input loads and system states， while structural parameters are treated as hyperparameters and optimized within the Gaussian process framework via an energy function and Markov Chain Monte Carlo （MCMC） sampling， thus achieving structural parameter identification. The effectiveness of the method in identifying structural parameters， input loads， and system states is validated through a numerical simulation of a 10-story shear frame， a vibration test on a 3-story frame， and comparisons with the augmented Kalman filter （AKF） and the dual Kalman filter （DKF）.

Abstract:In this paper， the on-site disaster investigation， tornado simulation test and finite element modeling analysis are carried out for the multi-span light steel industrial building that was seriously damaged in the Shengze tornado. Test parameters are the swirl ratio， the distance from the simulated tornado to the building model and the building orientation， and the finite element model focuses on the superstructure of the light steel industrial building. The wind load characteristics and structural response of the light steel industrial building under the EF3 tornado are systematically analyzed with several parameters， and the most vulnerable section of the industrial building under the tornado-induced wind load is given. The results show that under the action of tornado， the damage degree of the roof of different spans on the multi-span light steel industrial building is different， and the windward roof edge of the side span is the most seriously damaged section. The finite element analysis verifies that the tornado-induced wind pressure on this section is the largest， therefore the most unfavorable structural response is generated.

Abstract:The Hansbo’s non-Darcian flow model along with a nonlinear compression relationship was introduced to derive a one-dimensional nonlinear consolidation model for a double-layered soft soil foundation under linear loading while considering continuous drainage boundary conditions. An implicit finite difference method was subsequently employed for numerical solution. Numerical analyses indicate that， in the double-layered foundation， an increase in the Hansbo’s flow parameters will lead to a reduction in the rate of soil settlement. If the ratio of initial compression coefficients of the lower soil layer to the upper soil layer is large， the dissipation rate of pore pressure slows down， resulting in a decrease in the settlement rate of the foundation. Furthermore， an increase in the ratio of the initial permeability coefficients of the lower soil to the upper soil results in a slower dissipation rate of pore pressure in the upper soil. In addition， although a larger initial permeability coefficient ratio increases the soil settlement rate， it does not have a significant impact on the final settlement.

Abstract:In this paper， a BO-BiLSTM-Att surrogate model is proposed to efficiently calculate the excavation-induced deformation. The BiLSTM network is integrated with Att to capture temporal mapping relationships between input and output parameters. The Bayesian optimization （BO） is implemented to identify optimal hyperparameters， thereby further enhancing the model’s computational accuracy. Through a numerical case， the applicability and efficiency of BO in hyperparameter optimization are verified by comparison with PSO and GA. Subsequently， the surrogate model is trained by BO-optimized hyperparameters and its effectiveness in staged deformation calculation is demonstrated through the difference between predicted values and true values of test sets. Finally， the computational accuracy between BiLSTM network and BiLSTM-Att under identical hyperparameters is compared. It is shown that Att can improve the model’s computational accuracy. The proposed surrogate model has enhanced the computational accuracy （R²=0.967） and generalization capability （mean average error 0.875 mm） in testing scenarios.

Abstract:To address the poor allocation capability and weak coordination among crews during construction， a multi-agent proximal policy optimization （MAPPO）-based method for collaborative task allocation was proposed. A multi-agent construction simulation environment was first developed using Unity 3D. A Markov decision process model was then formulated， and a MAPPO-based framework with centralized training and decentralized execution was designed. Additionally， a valid action filtering （VAF） mechanism for construction tasks and a progress-based reward shaping （PBRS） method were proposed， enabling the efficient environmental exploration by agents and addressing the challenges of variable action spaces and sparse rewards in task allocation. Case study results demonstrate that the proposed method reduces the project duration by 14.8% and improves the labor utilization by 9.7% compared to conventional multi-agent methods. Furthermore， comparative analysis between MAPPO and independent proximal policy optimization （IPPO） demonstrates the necessity of collaborative decision-making for task allocation.

Abstract:To address the decision conflict and trajectory planning coupling issues arising from the spatio-temporal overlap of predefined intentions between vehicles during lane-changing scenarios on highways for intelligent connected vehicles （ICVs）， a collaborative trajectory planning method incorporating a three-stage screening process was proposed. Firstly， a quintic polynomial was employed to generate the combined trajectory solution space for both vehicles.Secondly， a hyperbolic tangent function was utilized to establish an elliptical safety boundary model for the initial trajectory screening. Thirdly， integrating the risk field theory with lateral and longitudinal impact intensity， a total risk function was constructed for the secondary theory screening. Finally， by combining vehicle dynamics constraints with a multi-objective lane-change cost function， overall trajectory combinations were evaluated to select the optimal combination. Two-vehicle lane-change simulation results demonstrate that under both normal and emergency lane-change scenarios， the proposed algorithm reduces lane-change costs compared to traditional risk assessment methods and effectively balances safety and comfort， which provides a reference for developing cooperative decision-making algorithms for connected vehicles.

Abstract:In this paper， a bi-level programming model is built for the train operation scheme optimization of multiple high-speed railway lines within a regional railway corridor， with the passenger assignment integrated into the planning process. The railway operator revenue maximization is set as the upper-level objective， while the passenger generalized travel cost minimization is adopted as the lower-level objective. Each OD demand is treated as a distinct commodity， and passenger flows are assigned using a multi-commodity flow integer programming model. A nested heuristic solution algorithm is designed： the upper-level problem is solved by simulated annealing combined with a feedback neighborhood search strategy based on seat load factors， and the lower-level model is solved using GUROBI. Case study results show that the optimized train operation scheme better matches passenger flow demand， with an average seat load factor of approximately 97%. Compared with the line-by-line independent optimization， the integrated corridor optimization shows a relatively small difference in ticket revenue and reduces variable cost by 18.4%.

Abstract:For a suburban railway line system with a double Y-shaped structure incorporating a shared section， formulation principles for the candidate train route set were established， and the line was divided into five sections based on the type of passing trains. Based on the passenger choice behavior analysis， passengers’ traveling between different sections was classified into three categories. An optimization model for cross-line train operation schemes on the suburban line was developed， with the objective of minimizing both enterprise operating costs and passenger travel time costs， and was solved using a genetic algorithm. Taking the Guanghua and Fangbai lines in Guangzhou as a case study， the validity of the model and the applicability of the algorithm were verified. The results indicate that the scheme in which both lines operate cross-line trains is generally superior.

Abstract:Aiming at the “strong port and weak railway” issue caused by the inefficient railway-waterway intermodal connectivity in current port-railway systems， an automated container terminal system based on air-rail connection （ACTAR） was proposed. By introducing a monorail and two-way operation system of air-rail collection and distribution， the operation process and efficiency of multimodal container transshipment were studied. Combined with the characteristics of container horizontal transportation operation， the joint scheduling optimization of track guided freight vehicle （TGV） and automatic guided vehicle （AGV） was conducted. A multi-objective linear integer programming model was built to minimize the maximum completion time and total delay time of the transportation system. The non-dominated sorting genetic algorithm Ⅱ （NSGA-Ⅱ） was used to solve the simulation system. The results show that the proposed model and algorithm can realize the parallel and seamless collaborative operation between TGV and AGV. There is an optimal threshold for the configuration number of TGV and AGV. Exceeding the limit will lead to a lower or even worse improvement in the operation efficiency of the transportation system. The number matching of the two is a key factor affecting the integrated scheduling optimization results of horizontal transportation equipment.

Abstract:To address the issues of strong dependence on mains electricity and poor deployment flexibility of conventional traffic monitoring devices， a low-frequency-driven rotary triboelectric nanogenerator （LDR-TENG） adapted to low-frequency， transient， and intermittent excitations was designed. Centered on a rotary triboelectric structure， a rack-pinion-mainspring-spring composite transmission system was constructed to convert transient vertical loads into sustained rotational excitations， thereby enhancing the energy harvesting and signal stability.Test results show that stable periodic signals can be generated at loading speeds as low as 5 mm·min^-1， with voltage characteristics linearly correlated to the loading speed （R2=0.998 87）. LDR-TENG was further embedded in an intelligent road stud for field testing， where the measured voltage features exhibited a 97.92% correlation with vehicle speed. A single-lane deployment scheme based on wheel track distribution achieved a vehicle trigger coverage rate of 76.47%.

Abstract:Under repeated wheel-rail loading， rolling contact fatigue cracks in rails propagate along the rail surface， exhibiting a distributed evolution process involving physical initiation， stochastic diffusion， and subsequent growth. To predict the stochastic process of crack evolution， a hybrid crack development prediction method integrating physics-based modelling and data-driven approaches is proposed in this paper. Random parameters associated with crack initiation and diffusion are incorporated into the physical model of crack evolution. These parameters are estimated using an approximate Bayesian inference algorithm， and neural networks are further employed for data training. On this basis， a relationship between crack diffusion and track operational conditions is established by integrating physical mechanisms with stochastic data-driven modelling. The results show that， for a given curve radius R， the crack diffusion rate λ is positively correlated with the daily gross tonnage Gd， and the effect of Gd on λ becomes significant when R<650 m. In contrast， Gd has only a weak effect on the initial crack inclination angle θ， whereas R is negatively correlated with θ. Moreover， a decrease in R or an increase in Gd significantly advances initial crack diffuse time T0.

Abstract:In this study， a dynamic coupling model integrating crop growth and soil water movement is developed， by combining the high-performance groundwater flow model SERGHEI-RE with the crop growth model WOFOST. Field observation data are utilized for model calibration and validation. Through this modeling framework， the effects of soil hydraulic properties and air temperature on water use efficiency （WUE） are investigated by constructing representative soil hydraulic datasets and future climate change scenarios. Simulation results demonstrate that elevated temperatures accelerate crop phenological development but concurrently reduce crop yield by 0.88%?2.92% and WUE by 7.75%?7.96% per 1 °C increase. Sandy loam soil can mitigate the WUE decline under warming conditions through regulating transpiration dynamics， showing the greater thermal adaptation potential. Time-lagged cross-correlation analysis reveals an asymmetric hysteresis relationship between the leaf area index （LAI） and soil moisture （SM）. The LAI-driven SM depletion responds rapidly （10?25 d lag）， whereas the SM-regulated LAI growth exhibits delayed responses （40?60 d lag）. This hysteresis is further modulated by soil types and temperature elevation. The air temperature and soil pore-size distribution are identified by random forest modeling and SHAP as dominant factors governing WUE.

Abstract:To address the limitation of storm water management model （SWMM） in accurately simulating lateral overflow discharge from open channels， a method for calculating open-channel lateral overflow was proposed. Based on the improved SWMM， a one-dimensional （1D）-two-dimensional （2D） coupled model was developed， enabling the inclusion of lateral flow exchange between 1D channels and the 2D ground surface within the coupling framework. The method achieved a fully integrated 1D-2D urban flood modeling system that simultaneously accounts for vertical and lateral flow exchange. A coupled simulation platform was established by integrating the improved SWMM with a self-developed 2D shallow water equation solver. The coupled simulation platform was validated using an actual urban waterlogging case， with HEC-RAS employed as a reference. The simulation results from the coupled platform show good agreement with those from HEC-RAS， with a relative error in water balance of less than 0.5%， an error in surface waterlogging volume of less than 5%， and a channel water level error within 0.03 m.

Abstract:In this study， based on particulate matter and ozone data from air quality monitoring stations in Shanghai near-sea， offshore and central urban area in 2023， the Pearson’s correlation coefficient and XGBoost model are used to investigate the effect of meteorological characteristics and sea-land winds on PM_2.5， PM₁₀ and O₃. The results indicate， pollutants exhibit clear seasonal variation， with the highest concentration of PM_2.5 in winter， while O₃ concentration peaks occurr from May to August， PM₁₀ concentrations are the highest in April in coastal areas and in December in the central urban area. The occurrence of sea-land wind days at each site is central city （17 d） < offshore （47 d） < near-sea （73 d）， sea-land wind days in offshore and near-sea areas are the most in summer， and the central urban area shows no significant seasonal variation. When the sea-land wind day occurs， particulate matter concentrations in coastal areas decrease， and O₃ concentrations increase. The impact of sea-land winds on pollutants in the central urban area is relatively minor. Meteorological characteristics significantly influence pollutants in different seasons， the precipitation is significantly negatively correlated with PM_2.5 and PM₁₀ concentrations in spring and winter， the temperature and sunshine duration are positively correlated with O₃ concentration， and the humidity is significantly negatively correlated with O₃ concentration. The feature importance analysis of the XGBoost model indicates that the wind speed， near-surface air pressure， and temperature are the dominant meteorological characteristics for PM_2.5 concentrations， the near-surface pressure contributes the most to PM₁₀ concentration at each site， with the highest contribution of 74.7% to the central city， and the temperature has the most significant effect on O₃ concentration at all sites. The simulation results of the XGBoost model show that PM_2.5 and O₃ are better， and the overall prediction accuracy shows the central city < offshore < near-sea.

Abstract:Initial rotor position estimation （IRPE） is crucial for the sensorless control startup and high-performance operation of automotive dual three-phase permanent magnet synchronous motors （DTP-PMSM）. Based on the established mathematical model of pulse voltage injection and the salient polarity of leakage inductance in the harmonic subspace， an IRPE method based on voltage pulse sequence injection in the harmonic subspace was proposed. Pulse sequences were injected into the harmonic subspace to suppress torque disturbance， and an overdetermined fitting algorithm was adopted to extract rotor position information from the current response. Bench test results verified the effectiveness of the proposed method. The high-precision initial rotor position identification was achieved and the torque disturbance was significantly reduced， thus resolving the inherent dilemma of traditional methods that fail to balance the estimation accuracy of initial rotor position and low noise and vibration performance.