Abstract:To address the issues of unbalanced load and system nonlinearity in quadrotor unmanned aerial vehicles(UAV)attitude control,a LSTM-MPC strategy is proposed by combining the advantages of long short-term memory(LSTM)neural network and model predictive control(MPC).LSTM neural network is used to predict attitude changes,enhancing the system's ability to anticipate errors.MPC is employed as feedforward control to dynamically optimize control inputs.The combination significantly improves system control accuracy.MATLAB simulation experiment on quadrotor UAV attitude control with unbalanced load shows that:compared to MPC strategy,the LSTM-MPC strategy reduces the root mean square error of tracking expected values for roll angle,pitch angle,and yaw angle by 13.33%,12.31%,and 11.11%respectively;compared to fuzzy PID strategy,it reduces by 14.05%,25.33%,and 23.81%respectively.Flight test is conducted using a branded F450 quadrotor UAV platform carrying a 0.6 kg load to test unbalanced load attitude control.The test result shows that the average errors between the actual output and expected values of the quadrotor UAV's roll,pitch,and yaw angles using the LSTM-MPC strategy are 3.91%,5.31%,and 1.10%,respectively,indicating that the LSTM-MPC strategy can effectively improve the flight stability of quadrotor UAV attitude control with unbalanced load.

Abstract:To improve the comprehensive road performance of in-situ recycled asphalt mixtures,this study investigates the impact of the amount of rejuvenator on the performance of recycled asphalt based on rheological performance parameters such as rutting factor,creep stiffness,creep rate,and critical temperature difference.It proposes a method for determining the quality ratio of rejuvenator to reclaimed asphalt based on the relationship between continuous high-temperature grades and continuous low-temperature grades.This method balances the high-temperature and low-temperature performance of asphalt to define the quality ratio of rejuvenator to reclaimed asphalt.Based on the determined quality ratio,asphalt mixtures are prepared,and the high-temperature and low-temperature performance of these mixtures is compared with those determined by the penetration method for the same quality ratio of rejuvenator to reclaimed asphalt.The results indicate that as the quality ratio of rejuvenator to reclaimed asphalt increases,the rutting factor of recycled asphalt measured by dynamic shear rheology gradually decreases,the creep stiffness measured by bending beam rheology decreases,the creep rate increases,and flexibility improves.The quality ratio of rejuvenator to reclaimed asphalt determined based on continuous high-temperature and low-temperature grades results in a slight decrease in the high-temperature performance of the recycled asphalt mixture,but the water-saturated Marshall residual stability increases by 9.4%,the freeze-thaw splitting tensile strength ratio increases by 7.9%,and the bending failure strain of the small beam increases by 5.5%,effectively balancing the high-temperature and low-temperature performance of the recycled asphalt mixture.

Abstract:To analyze the coupling coordination development level between low-carbon logistics and regional economy in Shandong Province,a multidimensional evaluation index system for low-carbon logistics and regional economy is constructed.The entropy weight method and a coupling coordination degree model are employed to assess the comprehensive development levels of low-carbon logistics and regional economy and their coupling coordination degree in Shandong Province from 2014 to 2023,identifying key factors influencing their coordinated development.Results indicate that the comprehensive development levels of both low-carbon logistics and regional economy in Shandong Province continuously increase from 2014 to 2023,with the low-carbon logistics composite score rising from 0.066 3 to 0.912 6 and the regional economy composite score growing from 0.032 1 to 0.994 6.The coupling coordination degree increases from 0.215 to 0.976,transitioning from moderate imbalance to primary coordination and further to high-quality coordination.The order degree of regional economy and low-carbon logistics shows a significant positive correlation,reflecting a highly synergistic interaction mechanism with sustained positive coordination trends.The main constraints for coordinated development are ecological level and industrial structure.Shandong Province can achieve low-carbon transformation of logistics and high-quality development of the regional economy by optimizing transport networks,promoting energy structure transition,advancing green logistics technologies,and developing low-carbon agriculture.

Abstract:To investigate the dynamic characteristics of underwater explosion bubbles under cylindrical shell structure,an experimental system for bubble pulsation and jet under cylindrical shell is designed.Bubbles are generated using an electric spark generation device,and bubble dynamic characteristic images are captured by a high-speed camera.Pressure sensors are employed to collect data on pressure load variations at the bottom wall of the cylindrical shell,and the influence of the dimensionless parameter γ on the peak pressure and duration of bubble loads are analyzed.Experimental results indicate that when γ is 0.25 and 0.50,the bubble shape is inverted mushroom-shaped to cumulous-shaped;when γ is 0.75 and 1.00,the bubble shape is columnar to cumulous-shaped;when γ is 1.25 and 1.50,the bubble shape exhibits egg-shaped to fan-shaped morphology.For 0.25≤γ≤1.50,the peak pressure of bubble shock wave load decreases with increasing γ,while the duration of shock wave load peak pressure increases with increasing γ;the duration of peak pressure from the first-cycle pulsation and jet load is greater than that of the second-cycle pulsation and jet load.For 0.25≤γ≤0.75,the peak pressures of both first-cycle and second-cycle pulsation and jet loads decrease with increasing γ;for 0.75<γ≤1.00,the peak pressures of both first-cycle and second-cycle pulsation and jet loads increase with increasing γ;for 1.00<γ≤1.50,the peak pressure of first-cycle pulsation and jet load decreases with increasing γ,while the peak pressure of second-cycle pulsation and jet load increases slowly with increasing γ.

Abstract:To address the issues of low additive utilization poor low temperature and fatigue performance in high modulus asphalt mixture,for direct application,a self-reactive high modulus asphalt and asphalt mixture are prepared using low-density ethylene-styrene copolymer-based high-modulus agents,and their technical performance is evaluated.Three kinds of self-reactive high modulus asphalt are prepared with rather lower content(6%,7% and 8%)of low-density ethylene-styrene copolymer as high modulus agent by pure mixing process,then the dispersion state and modification mechanism of high modulus agent are analyzed based on microscopic and physicochemical properties.The reasonable content of high modulus agent is selected based on the results of macroscopic performance test.Furthermore,the high modulus asphalt mixture with the preferred modifier content is prepared and its pavement performance is evaluated by laboratory tests.Results show that the modification process of self-reactive high modulus asphalt is a combination of the physical action of the high modulus agent and the chemical reaction between the asphalt binder and the modifier.The preferred high modulus agent content is 7%(by mass of the asphalt),which is 50%-80% of the modifier content in the dry-process.In this condition,the microscopic dispersion effect of the high modulus agent and the low-temperature performance of high modulus asphalt are desirable,and the PG high-temperature grade of high modulus asphalt is improved by one grade compared with that of the conventional high modulus asphalt.In addition to its superior high temperature performance,the high modulus asphalt mixture with optimum modifier content exhibited an 18%improvement in low-temperature destructive strain relative to conventional high modulus asphalt mixture with dry-process,while also achieving a 2 to 3 times extension in fatigue life.

Abstract:To fully leverage the wave-absorbing and heating properties of iron ore tailings and achieve high-value utilization of mining solid waste,this study incorporates iron ore tailings into cement concrete as coarse aggregates to investigate the damage patterns of iron ore tailing concrete under microwave irradiation.X-ray diffraction(XRD)phase analysis is performed on iron ore tailing particles to study the microwave heating mechanism.The temperature changes of iron ore tailing particles and concrete,as well as the ultrasonic pulse velocity(UPV)variations of iron ore tailing concrete before and after microwave irradiation,are analyzed.Scanning electron microscopy(SEM)is employed to observe the surface morphology and microstructure of the concrete before and after irradiation to investigate internal micro-damage.The results indicate that the iron-containing oxides in iron ore tailings possess superior wave-absorbing capabilities.The temperatures of both iron ore tailings and limestone increase with irradiation time,while the heating rate decreases;however,the temperature rise of iron ore tailings is larger than that of limestone.At the same irradiation time,the temperature of iron ore tailing concrete increases with increasing power.The wave-absorbing efficiency is highest at 6 kW.The temperature difference of the specimens before and after irradiation increases linearly at first and then slows down with increasing irradiation time.The UPV of the concrete decreases with increasing irradiation time and power.At the same power,the UPV attenuation rate increases rapidly initially and then slows down with increasing time.Under 0、2、4、6,8 kW power and 0～5 min irradiation,the UPV attenuation rate can reach up to 35.29%.After microwave irradiation,1～3 μm thermal stress cracks are generated in the interfacial transition zone between iron ore tailings and mortar and propagate along the interface.

Abstract:To predict the stability conditions of the support structure and achieve effective early warning,finite element software MIDAS CIVIL is used to model the pin-supported disc-type supports of the bridge,simulating the response characteristics of the support structure under single-point settlement and regional settlement conditions.Using in-situ testing methods on the bridge,a comparative analysis is conducted between the simulated data and the actual measurement data of the support deformation,demonstrating the feasibility of using angular change thresholds to monitor the preloading and pouring processes,and acquiring the deformation patterns at critical positions of the supports.The results show that the deformation data measured by the gyroscope closely match the simulated data,and the angular change threshold for construction process monitoring is feasible.Insufficient foundation bearing capacity leads to three different settlement conditions:first,when the single-point foundation bearing capacity is insufficient,forced displacements are applied to the four corner support rods at the bottom plane of the middle part of the support structure,resulting in significant angular changes at four measurement points,but none exceeds the warning threshold;settlement occurred at the position where the largest reaction force from the support is applied at the bottom of the support;second,when the foundation bearing capacity is insufficient in certain areas,forced displacements are applied to the middle,left,and right parts of the support structure,significantly affecting the key monitoring sections,with noticeable changes in local structural displacements;the angular changes remain within the permissible thresholds,and the reduction rates of displacement and angle during the pouring condition show minimal variation;third,when the foundation bearing capacity at the bottom of the members where each measurement point is located is insufficient,simultaneous node displacements are applied to the bottom nodes of the monitoring sections,with major settlement occurring in the regions around the bottom of the measurement points;the angular threshold increases with the increase in load,validating the effectiveness of the settlement identification and early warning method based on angular thresholds.

Abstract:To analyze the stress conditions of bridge deck panels with different stiffener forms in steel box girders,the software ANSYS is used to establish a comprehensive model and a spatial shell local model for a single box double-chamber continuous steel box girder with a span of(32+40+32)m.The effects of the main girder system and bridge deck system are obtained,and a comparative analysis is conducted on the structural effects of commonly used open section(inverted T-shaped)stiffeners and closed section(U-shaped)stiffeners in bridge deck panels.The results indicate that:1)The overall model reveals that the stress situation of the main girder system in the steel box girder,under basic combination effects and fatigue load effects,shows that the displacements and stresses of the bridge deck panels with inverted T-shaped stiffeners and U-shaped stiffeners are quite similar.2)Under local vehicle load,the longitudinal and transverse stresses of the inverted T-shaped stiffener bridge deck panel are more than twice those of the U-shaped stiffener bridge deck panel;the maximum longitudinal stress of the inverted T-shaped stiffener is greater than that of the U-shaped stiffener,while the vertical shear stress is smaller than that of the U-shaped stiffener and is lower than that of the bridge deck panel;the transverse and vertical stresses of the U-shaped stiffener steel box girder web are significantly increased compared to the inverted T-shaped stiffener,with a maximum increase of about 3.5 times.In design,adjustments should be made to the stiffness of the longitudinal and transverse stiffeners to ensure balanced stress in all components;the local deformation trends of the bridge deck panels in the inverted T-shaped and U-shaped stiffener models are similar,but the local deformation of the bridge deck panel under the inverted T-shaped stiffener model is larger than that of the U-shaped stiffener model,approximately 1.2 times the displacement of the latter.

Abstract:To address the problems of high scene complexity,difficult recognition,and high false detection rate in detecting abnormal behavior of object passing through subway station barriers,a detection method is proposed that data fusion based on light detection and ranging(LiDAR)and camera sensors.A voxel difference algorithm is used to process LiDAR point cloud data,divide the detection area into voxel units,and establish a trigger mechanism for object passing.An object locking algorithm is employed to fuse data collected by LiDAR and cameras,supplementing depth information of human key points and locking onto target of object passing through barriers.The spatial-temporal graph convolutional network(STGCN)is lightweight-modified to reduce model complexity and computation time.A temporal trend attention(TTA)model is introduced to enhance the extraction of spatial-temporal feature changes in postures of object passing through barriers,forming the TTA-STGCN model to calculate the confidence of behavior occurrence of object passing through barriers.Collecting data of object passing through barriers through laboratory simulation and on-site in subway stations.Detection performance evaluation metrics are established.Training,validation,and testing of STGCN,STGCN-MIN,and TTA-STGCN models are conducted.In the training phase,the accuracy of the TTA-STGCN model improved by 3.73%compared to the first two,the overall loss decreased by 66.00%.In the validation phase,the accuracy of the TTA-STGCN model improved by 3.89%and 0.68%compared to the first two respectively,the overall loss decreasing by 58.95%and 58.48%respectively.In the testing phase,the accuracy of the TTA-STGCN model improved by 3.15%compared to the first two,the overall loss decreasing by 42.85%and 44.40%respectively.Field experiments show that the TTA-STGCN model's accuracy improved by 2.99%and 3.49%compared to STGCN-MIN and STGCN models respectively,precision improved by 2.28%and 1.31%respectively,recall improved by 4.30%and 6.45%respectively,and F1 score improved by 0.033 5 and 0.040 4 respectively,demonstrating that the TTA-STGCN model significantly enhances the detection accuracy of behavior of object passing through barriers in specific subway station scenarios.

Abstract:To address issues such as large volume,heavy weight,and low adsorption force of permanent magnetic wheel in wall-climbing robot,the design aims to reduce the wheel mass and increase adsorption force.Combining lightweight design methods,the magnetic disks on both sides of the permanent magnetic wheel are designed by structural optimization.Electromagnetic field simulation software ANSYS Maxwell is used to model and simulate the permanent magnetic wheel,obtaining the magnetic field line distribution.By reducing the magnetic material in areas with sparse field lines and increasing it in dense areas,the wheel's adsorption force is enhanced.Through adjusting the main structural parameters of the magnetic disk and simulating the adsorption force changes between the permanent magnetic wheel and ferromagnetic wall surface,the optimal structural parameters for maximum adsorption force are determined.Simulation and physical tests verify the adsorption force of the optimized magnetic disk.Results show that the magnetic disk mass is 232.53 g,reduced by 16.24%;the simulated adsorption force between the permanent magnetic wheel and ferromagnetic wall is 487.65 N,increased by 20.74%;50 pull tests of the permanent magnetic wheel all exceed the simulated adsorption force,with an average of 493.83 N.The results show that after structural optimization,the mass and adsorption force indicators of the permanent magnet rotor meet the design objectives.

Abstract:To address the congestion problems caused by saturated and oversaturated traffic flow at urban road intersections and improve the efficiency of intersection traffic flow organization,an optimization method for traffic flow organization at pre-signal controlled intersections is proposed.Based on the channelization design of intersection approach lane and the coordination of main and pre-signal phase sequences,this method aims to improve the efficiency of left-turn and through traffic flows at intersection approaches and the utilization of intersection temporal and spatial resources.Two operation modes are designed for left-turn and through phases at approaches that continuous release mode and separate release mode.Experimental validation is conducted.Results indicate that:under different traffic volumes at intersection approach lane,pre-signal controlled intersections have smaller average signal cycle times,mean vehicle delays,and rates of increase in average vehicle delay with increasing approach lane traffic volume compared to traditional single-stop-line intersections,the separate release mode of pre-signal controlled intersections performs better than the continuous release mode in terms of average signal cycle times,mean vehicle delays,and rates of increase in average vehicle delay with increasing approach lane traffic volume;under the separate release mode of pre-signal controlled intersections during traffic flow saturated or oversaturated conditions on intersection approach lane,average vehicle delay time is minimized.

Abstract:To address the local optimum problem of the traditional artificial potential field(T-APF)algorithm and the limitations of dynamic obstacle avoidance in emergency collision avoidance scenarios for autonomous vessels,an enhanced artificial potential field(E-APF)algorithm based on time to closest point of approach(TCPA)and distance to closest point of approach(DCPA)is proposed.This algorithm reconstructs the repulsive potential field function,introduces a dynamic weight adjustment mechanism,and combines relative motion dynamics to design an adaptive repulsive direction strategy.Simulation results indicate that in static obstacle scenarios,the E-APF algorithm can identify collision risks earlier and plan better paths compared to the T-APF algorithm;in dynamic obstacle scenarios,it effectively increases safety distances and reduces steering angles,significantly improving the accuracy of obstacle risk assessment and collision avoidance decision-making.

Abstract:To accurately simulate the regenerative braking conditions of micro electric vehicles,a regenerative braking test system model is developed using MATLAB/Simulink.The simulation analyzed regenerative braking performance and calculated the regenerative efficiency of the drive motor under forced braking and gentle braking modes.The results show that when the accelerator pedal opening reaches 95%,the regenerative efficiencies under forced and gentle braking modes are 60.3%and 14.6%,respectively,which align closely with the target values of 60.0%and 15.0%.Additionally,higher braking deceleration and longer regenerative braking duration lead to greater recovered energy.The forced braking mode of the drive motor can significantly increase regenerative energy,thereby extending the driving range of micro electric vehicles.

Abstract:To reflect the actual state of the unsaturated three-phase medium of the subgrade,the subgrade is regarded as an unsaturated porous medium with incompressible solid and liquid phases,and a compressible gas phase.Based on the representation method of unsaturated porous media,considering the effects of inertia,viscosity,and mechanical coupling,a dimensionless one-dimensional wave equation for the unsaturated subgrade is established.Taking a type of non-homogeneous boundary as an example,the problem is transformed into an initial value problem of a second-order ordinary differential equation through boundary homogenization and the method of characteristic functions,and the state-space method is used to obtain the exact solution of the one-dimensional transient response of a single-layer unsaturated subgrade.The correctness of the results is verified through numerical examples.The characteristics of the one-dimensional transient response of the single-layer unsaturated subgrade under step and sinusoidal loads are analyzed.The results show that,unlike compressible saturated and unsaturated soils,there is only one type of compression wave in the incompressible unsaturated subgrade.As the liquid phase saturation increases,the wave speed and vibration amplitude of the compression wave decrease.The conclusion can provide a reference for the optimal design and disaster prevention of unsaturated subgrades under complex environmental loads.

Abstract:To measure smart logistics efficiency in the Beijing-Tianjin-Hebei Region,panel data on input indicators,output indicators,and influencing factors from 2013 to 2022 are utilized.The data envelopment analysis(DEA)model is applied to calculates technical efficiency,pure technical efficiency,and scale efficiency of regional smart logistics.The Tobit model is used to analyzes the impact intensity of economic development level,industrial factors,energy utilization rate,digital sales revenue,and internet development level on smart logistics efficiency.Recommendations for enhancing efficiency are proposed.Results indicate:during the period 2013 to 2022,Beijing′s smart logistics technical efficiency shows stage-wise leaps,continuously increasing from 2013 to 2015 and remaining efficient after 2016 except 2021;Tianjin achieves efficiency in 8 a,with inefficiency only in 2015 and 2019;Hebei achieves efficiency in 7 a,with inefficiency in 2015,2016,and 2020;low pure technical efficiency primarily causes inefficiency in Beijing while low scale efficiency dominates in Hebei;economic development level,industrial factors,and digital sales revenue exhibit significant positive effects on smart logistics efficiency,whereas energy utilization rate and internet development level show insignificant impacts.Recommendations include optimizing industrial structure,strengthening economic drivers,and advancing smart logistics development.

Abstract:To address issues such as poor stability,large trajectory tracking errors,and complex control strategies in the actual operation of two-wheeled self-balancing robot(TWSBR),the dynamics and kinematics of TWSBR are analyzed and a mathematical model is established.A forward control system is designed based on the linear quadratic regulator(LQR),and a steering control system is designed based on the sliding mode variable structure controller(SMVSC).Simulation experiments are conducted on the LQR forward control system and SMVSC steering control system.Simulation results show that:the actual steering angle of TWSBR reaches the desired steering angle after 1.2 s,the actual body tilt angle reaches the desired body tilt angle after 2 s,and the actual speed reaches the desired speed after 3 s.A trajectory tracking generator is designed,and a TWSBR trajectory tracking control system is constructed,and a TWSBR experimental platform is built.TWSBR trajectory tracking experiment is conducted.The experimental results show that the errors of TWSBR body tilt angle,body tilt angular velocity,steering angle,and steering angular velocity stabilize near 0 after 1 s,while the displacement error and speed error stabilize near 0 after 2 s.The fluctuation range of each state variable of TWSBR is small after stabilization.Considering factors such as sensor acquisition errors and motor drive precision,the results indicate that the TWSBR demonstrates good trajectory tracking performance,and achieves displacement tracking,speed tracking,body tilt angle tracking,and steering angle tracking.

Abstract:To investigate the shear strength and electrical resistivity characteristics of cement-stabilized soil under varying moisture contents,specimens were prepared with moisture contents of 7.7％,9.7％,11.7％,13.7％,15.7％,17.7％,and 19.7％,and a fixed cement content of 8％.Direct shear tests were conducted under vertical pressures of 100,200,300,and 400 kPa to examine the variation patterns of shear strength and resistivity responses under both unconfined and confined conditions.The results indicate that under unconfined conditions,both cohesion and internal friction angle at different moisture contents exhibit quadratic relationships with the initial resistivity.The fitted curves show high coefficients of determination,confirming the feasibility of assessing shear strength based on resistivity measurements.A nearly linear correlation is observed between shear strength and failure resistivity,where shear strength decreases as failure resistivity increases.With increasing vertical pressure,the peak shear stress of the specimens generally increased.As the moisture content increased,the peak shear stress first increased and then decreased.Under confined conditions,both initial and failure resistivity values exhibited negative exponential relationships with moisture content.As the vertical stress increased,the degree of reduction in failure resistivity decreased across different moisture contents.Under different moisture contents,the failure resistivity of the specimen generally decreases with increasing vertical pressure.

Abstract:To address the issues of settlement control,slope stability,and ecological protection during the construction of highways on soft soil foundations adjacent to rivers,based on the actual engineering case of the sixth section of the Tianjin-Shijiazhuang Expressway,a pile-net-filtering collaborative theoretical model is established.The model employs a pile-net structure formed by prestressed high-strength concrete(PHC)piles and reinforced cushion layers,combined with a new scheme for filtering concrete slopes,and is compared with traditional methods.The research results indicate that the new scheme effectively improves the bearing capacity of the foundation and the stability of the subgrade,with post-construction settlement not exceeding 52 mum(37％less than traditional methods)and differential settlement less than 2.0 mm/m;the filtering slope reduces pore water pressure by 55.6％,and the safety factor increases by 14.4％compared to traditional slope protection schemes;the life-cycle cost is reduced by 19.6％,construction time is shortened by 66.7％,carbon emissions are decreased by 27.6％,suspended particulate matter emissions are reduced by 78.2％,and the pH value returns to neutral(7.6).This scheme effectively resolves the three major conflicts of bearing capacity-settlement,water flow-soil interaction,and construction window limitations,providing an efficient,stable,and environmentally friendly solution for similar soft soil foundation projects adjacent to rivers.

Abstract:To analyze the traffic safety situation in complex navigable waters and assess the risk of vessel collisions,this study integrates multi-source data resources such as the automatic identification system(AIS)for vessels,port spatial distribution,and geographic information.The focus of the research is on the Pearl River Estuary,where the distribution of vessel types,sizes,and speeds in this water area is statistically analyzed.Geographic information data is utilized to analyze the spatial layout characteristics of the ports in this area.The navigation routes are divided into 24 time periods,and the water area is segmented into 9 parts.Based on line density analysis,the spatial characteristics of traffic flow and route distribution in the water area are studied,and the collision risk of vessels in key maritime traffic safety areas of the Pearl River Estuary is calculated and visualized.The results indicate that the vessel types in the Pearl River Estuary are diverse,with large sizes and high speeds;the main outbound navigation route from Guangzhou Port,the secondary outbound route from Guangzhou Port to the Dagang outbound route(passing through Lujing Bay),and the Shunde waterway to the Liansha Rong waterway are the three major routes,while the others are general routes.During different time periods,Route 3 has the highest number of routes,while Route 5 has the least.Route 3 is bimodal,with a higher number of vessels navigating during the noon and nighttime.Routes 2,4,6,and 7 are unimodal,and Routes 1,5,8,and 9 are stable regions with no significant peaks or troughs and little variation.In Route 3,the distribution of collision risk among vessels varies significantly,with higher collision risks present at the intersections of public channels and in the turning areas of the public channel.

Abstract:To study the stability of industrial sugarcane residue(mass fraction of 1.2％)-low alkali ecological cement(mass fraction of 7.0％)improved expansive soil slopes(hereinafter referred to as improved expansive soil slopes)under rainfall conditions,this research analyzes the differences between improved and unimproved expansive soils slope,as well as the intrinsic relationship between slope stability,deformation,and seepage.Based on soil parameters obtained from laboratory tests,numerical simulation software COMSOL Multiphysics is used to conduct large-scale slope rainfall intensity reduction simulations to explore the changes in effective saturation,plastic strain,displacement,and slope stability coefficient of the two types of soil slopes during rainfall.The results show that:1)After continuous rainfall,the overall change in effective saturation of the two types of soil slopes is small,with the effective saturation of the improved expansive soil slope greater than that of the unimproved expansive soil slope;under the condition of considering only the influence of rainfall,the effective saturation of the surface layer of the improved expansive soil slope at failure is greater than that of the unimproved expansive soil slope.2)Under rainfall conditions,plastic deformation in the unimproved expansive soil slope mainly occurs at the toe,while the plastic strain in the improved expansive soil slope concentrates at the crest,with the plastic strain of the improved expansive soil slope slightly larger than that of the unimproved expansive soil slope.3)After rainfall,displacements in both types of soil slopes occur at the crest,with the displacement of the unimproved expansive soil slope being about twice that of the improved expansive soil slope;within a relatively large displacement range,the area of affected soil in the unimproved expansive soil slope is larger,while the affected range of the improved expansive soil slope is limited to within 50 m around the plastic deformation,with the displacement of the unimproved expansive soil slope extending to the toe.4)Considering the rainfall intensity reduction method,U-shaped plastic strain zones appear within both types of soil slopes,with relative sliding occurring between the surface soil and deep soil;the plastic strain in the unimproved expansive soil slope is about three times that of the improved expansive soil slope;the stability coefficients of the unimproved expansive soil slope and improved expansive soil slope are 1.12 and 1.14,respectively,with the maximum displacement of the unimproved expansive soil slope being about four times that of the improved expansive soil slope.