Abstract:In this paper， large-eddy simulations （LES） were conducted to investigate the pollutant dispersion around a high-rise building， with different plume buoyancies produced by the density difference between ambient air and polluted gas being considered. The accuracy of the numerical method used in this paper was validated by comparing the simulated results with the wind tunnel experiment data. The effects of different plume buoyancies on the time-averaged concentration， concentration transportation， and concentration fluctuation were discussed. The positive buoyancy of the light gas significantly increases the vertical convection and turbulent flux in the wake region behind the building， enhancing the dilution of pollutant. In contrast， heavy gas has an opposite effect under the influence of negative buoyancy. In addition， the upward plume buoyancy further enhances the concentration fluctuation， while the downward plume buoyancy suppresses it to some degree.

Abstract:Large-eddy simulation was applied in the investigation of uniform flow past two square cylinders at a spacing ratio of P/B = 1.5 （P is the center distance of two cylinders and B is the side length of the cylinder） and an incidence angle of α = 0°to 90°. The aerodynamic forces， flow patterns， surface pressures， and flow fields were analyzed. The flow interference is significant for two square cylinders at a small spacing ratio. The variation of mean aerodynamic forces on the downstream cylinder is quite different from that on the upstream cylinder or a single square cylinder. In addition， a larger absolute value of lift force might occur at the downstream cylinder. The leading-edge separation mode （α = 0° to 10°）， separation-bubble mode （α = 20° to 30°）， attached-flow mode （α = 40° to 60°）， and gap-side separation-bubble mode （α = 70° to 90°） are summarized from the flow around two close square cylinders. A strong negative pressure is observed at the gap region for the attached-flow mode and gap-side separation-bubble mode. The strength of negative pressure in the near weak rises with incidence， reaches the maximum around α = 50°， and declines afterward. A single recirculation zone is observed in the weak of two cylinders at the lead-edge separation mode. However， two recirculation zones are recognized for the other three modes.

Abstract:A detailed comparative study was conducted to investigate the influence of different heights of central high-rise building and wind incident angles on the surrounding pedestrian-level wind environment by means of the wind tunnel test， Reynolds-averaged Navier-Stokes （RANS） and large-eddy simulation （LES） simulations simutaneously based on a standard urban building model proposed by AIJ （Architectural Institute of Japan）. The results show that the overall distributions of the wind speed ratio at the measuring points obtained from the RANS and LES simulations respectively are both consistent well with the wind tunnel test data， while the LES results are relatively better， which is reflected by the average error of LES being only half of that the RANS. Generally， the RANS simulation underestimates the pedestrian-level wind speed to some extent， especially it cannot appropriately reflect the wind acceleration phenomenon on the leewards of the building. The pedestrian-level wind speed is positively related with the height of the central building， and the building height over 100m will accelerate the pedestrian-level wind speed apparently by 1.6 times. However， as the height of the central building exceeds 150m （till 200m） ， the pedestrian-level wind speed will not increase further. As the wind incident angle varies from 0° to 90°， the so-called Venturi effect phenomenon will occur in the leewards and the corners of the central high-rise building. The pedestrian-level wind speed will accelerate to its highest level when the incident angle reaches 45°， which indicates that the oblique incident wind is the most unfavorable condition for considering the pedestrian-level wind environment situation around the high-rise building.

Abstract:Lattice Boltzmann method-based large-eddy simulation （LBM-LES） is a new method to solve turbulence problems in recent decades. However， improper time step settings may affect the simulation accuracy of LBM-LES. This paper first analyzed and summarized the impact of time step δt on the results of LBM-LES，theoretically. An oversized δt will cause compressibility error in the velocity field， while a too small δt can lead to the over-relaxation colliding mode， causing the numerical oscillation of velocity field. Subsequently， LBM-LES simulations of an isothermal indoor airflow case were conducted to discuss these errors quantitatively. The results show that a large δt leads to a sharp density change， and the velocity field in the regions where the Mach number （M） in the lattice Boltzmann unit exceeds 0.3 showing that there are obvious compressibility errors. Meanwhile， a too-small δt causes apparent numerical oscillations of both time-averaged and fluctuating velocities. This phenomenon is more significant when the grid resolution is higher. Therefore，it is suggested that δt should be small enough to ensure M<0.3 in the maximum velocity regions， based on which， a larger δt should be utilized to prevent numerical oscillations.

Abstract:In this paper， a numerical simulation of pollutant diffusion was conducted at five different locations and the results were taken as the measurement. The adjoint equation was used to calculate the simulated concentration of the sensors in time-variant flow field. The likelihood function was constructed by measurement and simulated concentration and the posterior probability of source parameters in time-varying flow field was calculated based on Bayesian inference. The results show that the errors of inversion of source parameters depend on the error between the measurement and the simulated concentration. When the distance between the source and the sensors is greater， the posterior probability of source parameters shows a wider distribution， indicating the larger uncertainty of the inversion result. When the source is closer to the sensors， the uncertainty of the inversion result is significantly reduced. In addition， the influence of measured data at different stages of pollutant diffusion in the process of inversion was also discussed. The inversion errors and the standard deviation of posterior probability are found to be smaller by using the measurement in the initial stage of diffusion than by using the data in the stable stage， but the improvement is not obvious.

Abstract:In this paper， the Nelder-Mead （NM） algorithm was combined with the robot active olfaction method to locate two types of time-variant contaminant sources， which are attenuated and periodic sources， respectively. First， the concentration fields of the contaminant sources were simulated through computational fluid dynamics （CFD） simulation. Then， the time-variant contaminant sources were located using the NM algorithm. The results show that the success rates of locating the two time-variant contaminant sources are both above 80 %. Three influencing factors such as the number of robots， robot response time， and the maximum number of robot search steps were discussed. The analysis indicates that in the case of 100 m² indoor space， time-variant sources can be best located when 5 robots are working on the site with a setting of 2 seconds in robot response and a maximum number of 50 robot search steps.

Abstract:Based on the shear-stress transport （SST） k-ω model， a “modified SST k-ω model” was proposed by modifying the model parameters， the source terms， and the turbulent viscosity. A computational fluid dynamics （CFD） simulation of atmospheric boundary layer （ABL） considering wind veering was performed by using this modified model. The self-sustainable method was discussed in detail including the precursor simulations and the main simulations. The research shows that the simulation results of the modified SST k-ω model are consistent with field measurement values. The physical quantities can be self-sustained well by applying the wind profiles obtained from the precursor simulations as inflow conditions in the main simulations.

Abstract:In this paper， a high-temperature steady-state tensile test study of S32001 duplex stainless steel was conducted， and the main mechanical property indexes such as initial modulus of elasticity， nominal yield strength， tensile strength and elongation after fracture， etc. at elevated temperatures were obtained. The applicability of the Rasmussen model and the Gardner model was studied using the test data. Based on the Rasmussen model， the calculation formula of the hardening index of S32001 stainless steel was proposed， and the constitutive relation expression at elevated temperatures was established， while the mechanical properties of S32001 stainless steel at elevated temperatures were compared and analyzed with other types of stainless steel and Q235B structural steel. The study shows that the yield strength and ultimate strength of S32001 stainless steel decreases with the increase of temperature， which is lower than 50 % at room temperature at 600 °C， but the material strength at elevated temperatures is significantly higher than that of S30408 stainless steel， which has more superior fire resistance properties. The results of this paper can be used in the study of structural fire performance and fire resistance design.

Abstract:For the need of engineering application and aseismic design of assembled connection between a staggered steel truss and a column， three kinds of joints were designed with different connection constructions. The hysteretic behavior of chord-web-column truss joints was investigated experimentally under cyclic axial force. It was found that the welded connection between diagonal web member and gusset plate was a weak link， liable to cracking. After the diagonal web member was damaged in the truss joint above， two of the three joints were selected to continue the tests on the hysteretic behavior of chord-column moment frame joints under cyclic bending moment. The performance such as failure feature， load bearing capacity， ductility and energy dissipation as well as influence of connection construction details was analyzed for these joints before and after the diagonal web member was damaged respectively. It is concluded that the truss joints subjected to cyclic axial force has a high loading capacity and initial rigidity， but a general ductility and energy dissipation ability. The chord-column connection configuration has no significant effect on the joint performance. The moment frame joints subjected to cyclic bending moment still has a high bending capacity and energy dissipation ability， but the effect of chord-column connection configuration on the joint performance is significant. In order to resist earthquake action， it is suggested that both the upper and lower flanges of H-section chord member in a staggered steel truss should be connected with the column reliably， which will cause the chord to transfer load to the column by means of bending resistance after the diagonal web member damages and then realize the function of a strong joint.

Abstract:The resonant frequency and the tilt of the offshore wind turbine foundation should be strictly controlled at natural cyclic load such as wind， wave and current， resulting in the fact that the characteristics of small strain modulus of soil should be considered in the design. The hardening soil model with small strain stiffness （HSS） parameters of typical marine sand in Yangjiang， Guangdong， including the effective stress strength parameters and stiffness parameters， were obtained by using conventional oedometer， GDS triaxial and bender elements. The relationships among stiffness parameters of marine sand were also analyzed. It is found that the measured small strain shear modulus G₀ is much smaller than the predictions made by empirical equations based on clean sands. The method to predict the HSS model parameters of marine sands in this area are furthermore suggested.

Abstract:Due to the vegetation-induced turbulence， the suspended sediment concentration in the rigid vegetated bed increased significantly， relative to the bare bed. The traditional sediment model which calculates the bed shear stress by mean velocity cannot simulate the phenomenon. Therefore， based on the physical modelling experiment of hydro-sediment dynamics in vegetated flows， a three-dimensional mathematical model of wave propagation with submerged rigid vegetation was established using Flow–3D in this paper to simulate the hydrodynamic characteristics and sediment suspension process under the influence of vegetation. Meanwhile， the original sediment model in Flow–3D was improved by modifying Shields number using turbulent kinetic energy. Compared with the experimental data， the model could accurately reproduce the decrease of mean wave velocity in the entire water column and increase at the top of the canopy， the enhanced turbulence caused by vegetation， and the two peaks of turbulent kinetic energy in the wave period. Using the improved sediment model considering the influence of vegetation wake turbulence on sediment movement， the accuracy of sediment suspension simulation in vegetated bed was significantly higher than the results from the original sediment model.

Abstract:A dynamics model of the virtual track train was developed based on the theories of vehicle， tire and nonlinear dynamics. The coupling between vehicles and the tire-road interaction were also considered in the virtual track train dynamics model. Meanwhile， the correctness of the dynamics model was verified by field tests. The dynamic load characteristics and road friendliness of the virtual track train in full running conditions between stations were investigated through theoretical analysis and numerical calculations. In addition， the effects of running speed， road grade and acceleration/deceleration were also analyzed. The results indicate that the longitudinal force of each tire in the full running condition is constant. The longitudinal force is equal in magnitude and opposite in direction in the traction and braking conditions， and is zero in the uniform condition. In the traction condition， the root means square （RMS） of the vertical dynamic load on the front axle of Vehicle 1 is the largest， while in the braking condition， the rear axle of Vehicle 3 is the largest. The road friendliness is better in the uniform condition compared with that in the traction condition and braking condition. The RMS of vertical dynamic load and road friendliness are all positively correlated with running speed， acceleration/deceleration， and negatively correlated with road grade. This paper can provide recommendations for the operation mode of the virtual track train to improve operation efficiency， and can provide guidance for the selection of asphalt pavement to reduce road damage.

Abstract:In this paper， a 3D （three-dimensional） laser scanner was used to obtain texture data. Surface geometric parameters and 2D-wavelet-based multiscale parameters were proposed as inputs. Different parameters and classifiers were tested on eight types of mixtures with the standard gradation curve. The result shows that the MLP （multi-layered perceptron） performs best with the combination of the surface and 2D-wavelet parameters. The MLP regression model was used to estimate the gradation curves， and the R² was 0.849. The ablation experiments were used to analyze the contribution of different parameters. It is found that the 2D-wavelet energy parameters have more significant effects.

Abstract:A wear prediction model groove-shaped rail of modern trams is established， which fully considers the wheel-rail coupling effect of independent rotating wheel. The validity of the model was verified by comparison with the results of relevant literature. The characteristics of the rail wear of modern trams are analyzed by the proposed wear model. A profile optimization model of the rail based on augmented Gaussian radial primary function （AGRPF） is established， which is solved by using sequential quadratic programming （SQP） and multi-objective optimization principle. The results show that the main wear position of the outer rail is at the gauge angle of Y=30 mm， and the wear positions of the inner rail are at the top of the rail and the guard rail. Compared with the initial rail profile， the optimized effect of the rail profile is 15 % to 30 %， which can effectively reduce rail wear， especially side wear， increase wheel-rail contact area and reduce wheel-rail contact stress. The research results can provide a useful reference for the selection and maintenance of tram rails.

Abstract:Based on the technical problem and difficulty of the randomness and arbitrariness in the selection of the current train driving speed curve， this paper constructed the structure of the energy-saving driving assistance system， and designed the system information interaction mode. based on which， the energy-saving model algorithm of train running at regular time is established. The method of genetic algorithm model based on cruise speed （CS） was innovatively established， and the energy saving model algorithm of train running at regular time was calculated and analyzed. The optimal speed curve was solved by encoding the energy consumption and CS. The algorithm proposed in this paper does not need to predetermine the sequence table of working conditions， increases the degree of freedom of model optimization， and makes the overall train speed curve appear more stable. The results of simulation experiments verify the effectiveness of the proposed method， and the advantages of the proposed model are proved by comparing with the examples in typical literatures. The optimal speed curve solved in this paper can be finally transmitted to the on-board terminal of driver assistance system（DAS） through driver assistance information for display， which can be used for driving guidance and energy saving effect of train driving.

Abstract:In order to study the friction and wear performance of sliding pair of incremental launching equipment， a wear test-bed which can simulate the working process of incremental launching equipment is proposed. Tribological tests for the sliding pair composed of acrylonitrile butadiene styrene （ABS） and molybdenum disulfide （MoS₂） reinforced polytetrafluorethylene （PTFE） and 0Cr18Ni9 stainless-steel were conducted on the wear test-bed. At different loads without lubrication， the variation of friction coefficient of friction pair was discussed and the wear mechanism of the friction pair was revealed. The microscopic and chemical composition of the worn surface for the sliding pairs were discussed by using scanning electron microscope （SEM） and energy dispersive spectrometer （EDS）. The results show that， with the increase of sliding times， the friction coefficient of the sliding pair first increases， then decreases， and finally stabileizes. Moreover， the initial value of friction coefficient at heavy load is larger than that at light load， but the final stable value is smaller than that at light load. At light load， the main wear mechanism is abrasive wear and adhesive wear. However， at heavy load， the main wear mechanism is adhesive wear and fatigue wear.

Abstract:From the perspective of non-uniform sound field offset， the mechanism of pure-tone sound-amplitude large change with wind speed which measured outside flow field of automobile wind tunnel was studied. A wind tunnel test was conducted with a controllable sound source on the automobile surface as the research object. The amplitude fluctuation of pure tone sound pressure level at different wind speeds was obtained by microphone， and the non-uniform distribution of sound field was observed by microphone array. Based on the point source interference theory， the non-uniform sound field was simulated and reconstructed， and the influence of each reflecting wall was analyzed parametrically. The acoustic field offset caused by shear layer refraction was deduced according to the geometrical acoustic theory， and the trend of pure tone amplitude variation at different test wind speed was predicted. The results show that the deviation of the non-uniform sound field caused by the interference is the fundamental reason for the amplitude fluctuation of the pure sound in the wind tunnel， and the reconstruction method of acoustic field can effectively reduce the error.