交通运输工程学报

Review on cement concrete internally cured by SAP

SHEN Ai-qin, YANG Jing-yu, GUO Yin-chuan, QIN Xiao, LI Peng

Abstract: The material properties of a superabsorbent polymer (SAP) were analyzed. Key parameters of mix ratio designs for SAP internally curing concrete were evaluated. A design method for internally cured concrete was developed. The water transmission mechanism of SAP internally curing concrete was investigated from the perspective of the water absorption and water release behavior of the SAP and the hydration characteristics of concrete. The shrinkage and crack resistance, mechanical properties, and durability of SAP internally curing concrete were examined. Its performance enhancement mechanism was explored by considering the characteristics of the interfacial transition zone, hydration products, and pore structure. The engineering applications of SAP internally curing concrete at home and abroad, as well as future research directions and application prospects, were identified. Analysis results show that the principle of SAP internally curing concrete relies on its water absorption and release characteristics. However, there is some variability in the performance of internally cured cement concrete because of the differences in SAP performance and concrete mix ratio and other factors. SAP, as a result of osmotic pressure and ionic concentration, is able to release water over time to replenish the loss of water inside the concrete, reduce the early heat of hydration, and enhance the later hydration. The properties of SAP internally curing concrete are affected by its particle size and admixture amount, and additional water quantity. SAP can effectively suppress the self-shrinkage and drying shrinkage and enhance the mechanical properties of concrete when all the parameters are suitable. SAP can also promote the hydration reaction, generate more hydration products, fill pores of concrete, enhance the compactness of concrete, refine the pore structure, and break off the connected pores, thereby improving the durability of concrete, including its frost resistance and impermeability. The reswelling ability of SAP can block concrete cracks and generate hydration products, such as CaCO₃, to enable concrete to self-heal. The curing effects of SAP can enhance the adhesion between cement stones and aggregates, reduce or even eliminate the microcracks in the interfacial transition zone, and improve the strength of the interfacial transition zone. The SAP internally curing concrete can be successfully applied to bridge deck integrative layers, cross diaphragms, wet joints, bridge piers, tunnel secondary linings, etc. with an excellent anti-cracking effect. 46 figs, 137 refs.More>

2021, 21(4): 1-31. doi: 10.19818/j.cnki.1671-1637.2021.04.001

FullText HTML (656) PDF (41675KB) (120)

Review on detection and prediction methods for pavement skid resistance

TAN Yi-qiu, XIAO Shen-qing, XIONG Xue-tang

Abstract: The relevant achievements and progress of skid resistance performance of pavements were systematically reviewed based on three aspects: mechanical mechanism, detection methods, and prediction models. The friction mechanism of pavement skid resistance was introduced based on the traditional Coulomb friction law, and the factors influencing the skid resistance were summarized based on road surface, tire, and contact environment. The direct and indirect measurement methods of skid resistance were evaluated, and the difficulties of road-surface texture detection and test data preprocessing methods were analyzed. The advantages and disadvantages of the skid resistance prediction methods, including traditional empirical statistical models, mechanical models, and machine learning, were compared and analyzed. Analysis results show that many factors influence the skid resistance of pavements, and it is difficult to describe the mechanical behavior of the third body between rubber and rough surface. Thus, further investigations are required to reveal the friction mechanism toward the contact interface with the lubrication medium.For the single function and high cost of the direct detection of skid resistance, surface texture detection using automatic high-speed noncontact measurements will be more in line with future intelligent integrated requirements. However, high-precision and large-range detection and data cleaning are still bottlenecks that need to break through. Compared to the existing prediction models, the tire-pavement contact characteristics are weakened by the empirical statistical model and machine learning, resulting in a lack of scalability in the prediction model. The implementation of the finite element simulation model method is expected to reveal the friction mechanism under complex physical fields to develop a more precise and efficient model for predicting the skid resistance of pavements. 1 tab, 13 figs, 89 refs.More>

2021, 21(4): 32-47. doi: 10.19818/j.cnki.1671-1637.2021.04.002

FullText HTML (983) PDF (7461KB) (263)

Review of convolutional neural network and its application in intelligent transportation system

MA Yong-jie, CHENG Shi-sheng, MA Yun-ting, MA Yi-de

Abstract: From the perspectives of feature transmission mode, spatial dimension and feature dimension, the improvement directions of convolution neural network structure in recent years were reviewed. The working principles of the convolution layer, pooling layer, activation function and optimization algorithm were introduced, and the recent developments of pooling methods in terms of value, level, probability, and transformation domain were summarized. The comparison of some representative activation functions, and the working principle and characteristics of the gradient descent algorithm and its improved and adaptive optimization algorithm were given. The application and research status of convolutional neural network in intelligent transportation fields such as license plate recognition, vehicle type recognition, traffic sign recognition, and short-term traffic flow prediction were reviewed. The convolutional neural network algorithm was compared with the support vector machine, differential integrated moving average regression model, Kalman filter, error back propagation neural network, and long-term and short-term memory network algorithms from the advantages and disadvantages and main application scenarios in the field of intelligent transportation. The issues of poor robustness and poor real-time performance of convolutional neural network in the field of intelligent transportation were analyzed. The development trend of convolutional neural network was evaluated in terms of algorithm optimization, parallel computing, and supervised learning to unsupervised learning. Research results show that the convolutional neural network has strong advantages in the field of vision. It is mainly used for traffic sign, license plate, vehicle type recognition, traffic event detection, and traffic state prediction in intelligent transportation system. Compared with other algorithms, the convolutional neural network can extract more comprehensive features. It can effectively improve the recognition accuracy and speed and has great application value. The convolutional neural network will bring new breakthroughs to intelligent transportation in the future through the optimization of network structure, the improvement of algorithm and computing power, and the enhancement of benchmark data sets. 5 tabs, 3 figs, 146 refs.More>

2021, 21(4): 48-71. doi: 10.19818/j.cnki.1671-1637.2021.04.003

FullText HTML (775) PDF (1523KB) (405)

Shaking table test of liquefaction resistance of group piles under strong earthquake

FENG Zhong-ju, ZHANG Cong, HE Jing-bin, LIU Chuang, DONG Yun-xiu, YUAN Feng-bin

Abstract: To study the specific manifestation of the improved basic liquefaction resistance of group pile foundation in comparison with that of single-pile foundation under strong earthquake, considering the Haiwen Bridge Project in Hainan Province as an example, a shaking table model test was adopted to examine the differences in foundations comprising one, four, and six piles. The differences in the time-history responses of pore pressure ratio in saturated fine sand, pile acceleration, and bending moment under three different working conditions, and their relationships were analyzed. The results indicate that the liquefaction occurs under all three working conditions and a ground motion of 0.35g. The time when the pore pressure ratio begins to increase and that when the ratio becomes stable in the deep layers of saturated fine sand lag behind those in the shallower layers. The time required for the complete liquefaction of the foundation with six piles is 4.41-4.82 s longer than that for the foundation with four piles. The time required for the complete liquefaction of the foundation with four piles is 4.00-4.42 s longer than that for the single-pile foundation. With more piles, the maximum pile acceleration and its amplification factor in the saturated fine sand at the same depth decrease gradually, and the maximum pile acceleration gradually lags behind. In addition, as the pore pressure ratio increases, the pile acceleration decreases gradually. The maximum bending moment of the foundation with six piles is 25.95%-43.50% smaller than that of the foundation with four piles. Similarly, the maximum bending moment of the latter is 28.80%-33.10% smaller than that of the single-pile foundation. The maximum bending moment of the single-pile foundation appears 1.22-1.27 s earlier than that of the foundation with four piles, whereas that of the latter appears 0.66-0.72 s earlier than that of the foundation with six piles. Furthermore, the bending moment of pile gradually attenuates as the pore pressure ratio increases, indicating that the saturated fine sand provides softening and damping effects before liquefaction. In summary, the liquefaction resistance of the foundation with six piles is better than those of the foundations with four piles and one pile. Thus, in the antiseismic design of pile foundations for liquefaction-prone soil layers, the liquefaction resistance of foundations can be improved by using group pile foundations. 10 tabs, 12 figs, 32 refs.More>

2021, 21(4): 72-83. doi: 10.19818/j.cnki.1671-1637.2021.04.004

FullText HTML (259) PDF (12126KB) (44)

Bearing characteristics of cast-in-place piles with different hole-forming methods before and after grouting in loess area

ZHOU Zhi-jun, XU Tian-yu, XU Fu, CHEN Chao-ran, BAI Yang

Abstract: Static load tests on manual digging piles (MDPs), rotary drilling piles (RDPs), and impact drilling piles (IDPs) before and after grouting were performed to investigate the effect of hole-forming method on the bearing characteristics of cast-in-place piles under post-grouting. The effect of hole-forming method on the final grouting quantity was analyzed. The improvement effects of pile settlement, ultimate bearing capacity, pile side resistance, and pile end resistance before and after grouting under different hole-forming methods were evaluated. The influence mechanisms of different hole-forming methods on the grouting process and bearing characteristics of the pile were examined. The methods for calculating the ultimate bearing capacity of post-grouting pile recommended by the Technical Code for Building Pile Foundations (JGJ 94—2008) were verified by comparing different hole-forming methods and the upper return height of the slurry. Research results show that the actual final grouting quantities range from large to small in the order of RDPs, MDPs, and IDPs when the theoretical grouting quantities are the same. After grouting, the improvement effects of settlement characteristic range from positive to negative in the order of IDPs, RDPs, and MDPs. The increasing ranges of the ultimate bearing capacity are from large to small in the order of IDPs, RDPs and MDPs. The pile side resistance improves within 12 m above the pile end after grouting, and the increasing ranges are in the order of IDPs, RDPs, and MDPs. The proportions of pile end resistances to the total loads of MDPs, RDPs, and IDPs increase by 17.05%, 12.23%, and 15.10%, respectively, indicating significant improvements in the characteristics of the end-bearing friction pile. The enhanced coefficients of side resistances and end resistances of MDPs and RDPs are close to those recommended in the Technical Code for Building Pile Foundations (JGJ 94—2008), whereas those of the IDPs are slightly different. Therefore, the hole-forming method of cast-in-place pile significantly influences the selection of post-grouting parameters and the bearing characteristics of cast-in-place piles. 6 tabs, 7 figs, 31 refs.More>

2021, 21(4): 84-93. doi: 10.19818/j.cnki.1671-1637.2021.04.005

FullText HTML (210) PDF (10017KB) (30)

Computational model of outer-circle and inner-ellipse shield tunnel lining structure

GU Shuan-cheng, SUN Guan-lin, SU Pei-li

Abstract: In order to adapt the metro tunnel to the inhomogeneity of stratum load, referring to the structural form of special-shaped shield tunnel lining and taking into account the factors such as production and construction, the outer-circle and inner-elliptic shield tunnel lining structure was proposed, ensuring that the stiffness of the most unfavorable position of the lining structure meets the safety requirement, appropriately reducing the stiffness of the rest of the lining structure, and taking advantage of material properties. The flexibility coefficient and free term of outer-circle and inner-elliptic shield tunnel lining were solved by the stiffness step discounting method, and the calculation model of the lining was established. Referring to the actual engineering geological condition, the internal force distribution characteristics of outer-circle and inner-elliptic shield tunnel lining were studied. The safety of outer-circle and inner-elliptic shield tunnel lining was evaluated by the Code for Design of Railway Tunnel (TB 1003—2016). Calculation results show that compared with the equal stiffness shield tunnel lining, under the same loading condition, the outer-circle and inner-elliptic shield tunnel lining reduces the bending moments at the top and bottom of the lining structure, and transfers the maximum bending moment and maximum axial force to the arch waist, which simplifies the safety test by focusing on whether the internal force at the arch waist of the lining structure can meet the safety condition during the test. In terms of stability, the safety factors of equal stiffness shield tunnel lining are 3.07, 18.05 and 2.45 at the vault, arch shoulder and arch waist, respectively. The safety factors of outer-circle and inner-elliptic shield tunnel lining are 2.79, 14.86 and 2.21 at the vault, arch shoulder and arch waist, respectively, and slightly lower than those of the equal stiffness shield tunnel lining, but still greater than the minimum value 2.0 stipulated by the safety checking requirement, giving full play to the material characteristics of concrete. In terms of internal space, when the outer-circle and inner-elliptic shield tunnel lining and the equal stiffness shield tunnel lining have the same outer radius, the internal space area of the equal stiffness shield tunnel lining is 22.9 m², and the value of the outer-circle and inner-elliptic shield tunnel lining is 23.76 m² and significantly larger than that of the equal stiffness shield tunnel lining. Therefore, the area and utilization rate of internal space of the proposed lining increase without expanding the outer radius. 8 tabs, 11 figs, 31 refs.More>

2021, 21(4): 94-105. doi: 10.19818/j.cnki.1671-1637.2021.04.006

FullText HTML (171) PDF (4773KB) (20)

Lateral foundation reaction coefficient of feet-lock pipe in loess stratum

CHEN Li-jun, CHEN Jian-xun, LUO Yan-bin, WANG Chuan-wu, HU Tao-tao

Abstract: A simulated loading test of a feet-lock pipe in the Malan loess stratum was conducted. The proportional coefficient of lateral foundation reaction coefficient of the feet-lock pipe was obtained through the inverse calculation. The variations in the proportional coefficient of lateral foundation reaction coefficient with the vertical load and displacement at the end of the feet-lock pipe were studied. The influence of installation angle of feet-lock pipe on the lateral foundation reaction coefficient was analyzed. The regression equation of the proportional coefficient of lateral foundation reaction coefficient with respect to the vertical displacement at the end of feet-lock pipe was fitted. The upper and lower limits of the proportional coefficient of lateral foundation reaction coefficient were determined. The proportional coefficients of lateral foundation reaction coefficient of feet-lock pipe with arbitrary angles in the range of 0°-30° were acquired through the interpolation calculation. Research results show that the proportional coefficient of lateral foundation reaction coefficient increases rapidly as the vertical load and displacement at the end of feet-lock pipe increase in the early stage of loading. This occurs because the loess around the feet-lock pipe is compacted continuously. When the loess around the feet-lock pipe enters the progressive failure stage, the proportional coefficient of lateral foundation reaction coefficient decreases continuously as the vertical load and displacement at the end of feet-lock pipe increase. The proportional coefficient of lateral foundation reaction coefficient of a feet-lock pipe installed at an angle of 30° exhibits significantly smaller variations with the vertical load and displacement compared with that of a feet-lock pipe installed at an angle of 0°. In the design and calculation of feet-lock pipe of tunnels in Malan loess stratum, when a feet-lock pipe with a diameter of 51 mm is installed at an angle between 0° and 30°, the proportional coefficient of lateral foundation reaction coefficient ranges from 66.9 to 296.1 MN·m^-4. The vertical load that a single feet-lock pipe with a diameter of 51 mm can bear is between 8.63 and 11.87 kN. 2 tabs, 13 figs, 31 refs.More>

2021, 21(4): 106-115. doi: 10.19818/j.cnki.1671-1637.2021.04.007

FullText HTML (255) PDF (11824KB) (19)

Calculation method of UHPC local compressive bearing capacity

QIU Ming-hong, SHAO Xu-dong, LIU Qiong-wei, YAN Ban-fu, LI Pan-pan, HUANG Zhong-lin

Abstract: To reasonably calculate the local compressive bearing capacity of UHPC members, the local compression test database of UHPCs with and without indirect reinforcement was developed. Based on the database, the calculation formulae of local compressive bearing capacity in codes of NF P 18-710, CECS 38: 2004, DBJ 43/T 325—2017, and JTG 3362—2018 were analyzed and evaluated. A UHPC local compression correction factor and an indirect reinforcement influence factor were proposed to analyze the effects of concrete strength and steel fibers used in the UHPC local compression test database. Subsequently, the local compressive bearing capacity calculation formula in JTG 3362—2018 was modified. Research results indicate that the average ratios of experimental local compressive bearing capacities of UHPCs without indirect reinforcement to the calculated values by UHPC codes of NF P 18-710, CECS 38: 2004, DBJ 43/T 325—2017, and JTG 3362—2018 are 0.97, 0.81, 1.33, and 1.09, respectively. The average ratios of experimental local compressive bearing capacities of UHPCs with indirect reinforcement to the calculated values by CECS 38: 2004, DBJ 43/T 325—2017, and JTG 3362—2018 are 0.91, 1.31, and 1.13, respectively. Therefore, the calculation formulae of concrete local compressive bearing capacity in different codes do not completely reflect the influences of concrete compressive strength and steel fibers. Similarly, the local compressive bearing capacity calculation formulae of indirect reinforcement do not completely reflect the effects of constrained area ratio, concrete compressive strength, and steel fibers. NF P 18-710 can better predict the local compressive bearing capacity of UHPC without indirect reinforcement. CECS 38: 2004 overestimates the local compressive bearing capacity of UHPC, thereby diverging the estimated bearing capacity of indirect reinforcement. As for DBJ 43/T 325—2017 and JTG 3362—2018, their calculation results are relatively conservative. The average ratios of the experimental local compressive bearing capacities to the predicted values by the modified JTG 3362—2018 formula are 1.00 and 1.04 for UHPCs with and without indirect reinforcement, respectively, and the standard deviations in both cases are less than 0.20. Therefore, the modified formula in JTG 3362—2018 can better predict the local compressive capacities of UHPCs with and without indirect reinforcement, and thus it can provide a reference for the design specifications of domestic UHPC bridge structures. 1 tab, 13 figs, 31 refs.More>

2021, 21(4): 116-129. doi: 10.19818/j.cnki.1671-1637.2021.04.008

FullText HTML (364) PDF (12102KB) (33)

Comparison of vertical vortex-induced vibration characteristics between semi-open girder and separated edge-boxes open girder

DUAN Qing-song, MA Cun-ming

Abstract: Wind tunnel tests on sectional models with a scale ratio of 1∶50 were performed to comprehensively investigate the vortex-induced vibration characteristics and associated mechanisms of open girders with different cross sections. The vortex-induced vibration characteristics of semi-open girder and separated edge-boxes open girder were analyzed and compared. The influence factors, including the equivalent mass, wind attack angle, and damping ratio, were considered. In addition, the Strouhal numbers of the two girder cross sections were computed. Based on the linear and nonlinear theories, the vertical vortex-induced vibration amplitudes of real bridge girders were calculated. A two-dimensional numerical simulation model was established, and the accuracy of the numerical simulation method was verified. Then, the instantaneous vorticity contours and mean streamline structures around the two girder cross sections were compared. Analysis results show that at the wind attack angles of 3° and 5°, the vortex-induced vibration is observed for both girders, and there are two vortex-induced vibration regions. The maximum amplitude in the second vertical vortex-induced vibration region is significantly larger than that in the first one. The vertical vortex-induced vibration amplitude at an wind attack angle of 5° is 75% larger than that at an wind attack angle of 3°. When the wind attack angle is 5° and damping ratio is 0.8%, the maximum vertical vortex-induced vibration amplitude of separated edge-boxes open girder is 28% larger than that of the semi-open girder. The maximum vertical vortex-induced vibration amplitude decreases almost linearly as the Scruton number increases. For the same Scruton number, the vertical vortex-induced vibration amplitude of separated edge-boxes open girder peaks at a wind attack angle of 5°, whereas the vertical vortex-induced vibration amplitude of semi-open girder is at its minimum at a wind attack angle of 3°, indicating that the larger the positive wind attack angle, the blunter the cross section of the girder, and the worse the vortex-induced vibration characteristics. When the wind attack angle equals 5°, the separated edge-boxes open girder is blunter, causing increased air fluid separation. There are vortexes above the girder deck and at the openings of the girder formed by the incoming wind, the inclined web and wind fairing may break the large vortexes at the openings into several smaller vortexes with similar sizes, thus optimizing the vortex-induced vibration of girders. 3 tabs, 9 figs, 31 refs.More>

2021, 21(4): 130-138. doi: 10.19818/j.cnki.1671-1637.2021.04.009

FullText HTML (158) PDF (10129KB) (20)

Fire resistance performance and design method of steel-concretecomposite continuous curved box girders

SONG Chao-jie, ZHANG Gang, HE Shuan-hai, KODUR V K, HUANG Qiao, LI Xu-yang

Abstract: As a strategy to improve the fire resistant performance of steel-concrete composite continuous curved box girders, a three-span steel-concrete composite continuous curved box girder was selected as a research object to establish a two-stage three-dimensional nonlinear analytical model under fire by employing the commonly used finite element software ANSYS. Based on the existing thermal-structural coupled analytical method, the developed model considered the radiation heat transfer in the cavity of steel box girder and the contact boundary conditions at the interface between the top flange of steel box girder and the concrete slab. The prediction results obtained by the model were compared with the experimental data to verify the model's reliability. The established model was used to conduct a parameter sensitivity of mid-span deflection of the steel-concrete composite continuous curved box girder under different longitudinal fire exposure positions, fire intensities, and load levels. The decay laws of ultimate bearing capacity and stiffness of the steel-concrete composite continuous curved box girder was studied. With the mid-span deflection under fire used as the evaluation indicator, a fire resistant design method of steel-concrete composite continuous curved box girders was proposed. Research results show that the deflection of the outer edge of steel-concrete composite continuous curved box girder is greater than that of the inner edge under the symmetrical fire and structural load, and this effect is more significant with greater loads and more severe fire. The stiffness decreases faster than the ultimate bearing capacity under a large burned area such as that resulting from a fuel tanker fire. Compared with the ultimate bearing capacity and stiffness of steel-concrete composite continuous curved box girder under normal temperature, the ultimate bearing capacity and stiffness reduce to 29% and 14%, respectively, when the side span is exposed to fire for 16 min, and they further reduce to 31% and 22%, respectively, when the middle-span is exposed to fire for 28 min. In the fire resistant design of steel-concrete composite continuous curved box girders, improving the stiffness of outer steel box girder under fire is necessary. Increasing and widening the longitudinal stiffeners of the bottom plate of outer steel box girder can reduce the mid-span deflection difference between the inner and outer steel box girders by 23% and 30%, respectively, when the side span is exposed to fire for 20 min, and by 22% and 27%, respectively, when the middle-span is exposed to fire for 32 min. 1 tab, 15 figs, 31 refs.More>

2021, 21(4): 139-149. doi: 10.19818/j.cnki.1671-1637.2021.04.010

FullText HTML (294) PDF (9928KB) (44)

Propagation characteristics of initial compression wave in cave and portal micro-pressure waves characteristics when 600 km·h^-1 maglev train entering tunnels

MEI Yuan-gui, LI Mian-hui, HU Xiao, DU Jun-tao

Abstract: Based on a three-dimensional numerical simulation method, a one-dimensional compressible unsteady non-isentropic flow model and an improved generalized Riemann variable characteristic line method were developed. The initial compression wave propagation in the tunnel and the micro-pressure wave characteristics at the portal (default exit) of the tunnel when the tunnel entrance without and with an opening buffer structure were investigated. Analysis results show that compared to the nonbuffer structure at the tunnel entrance, the maximum pressure gradient of the initial compression wave generated by setting the opening buffer structure decreases by 67.56%. During the propagation of the initial compression waves in the tunnel, intensification first occurs, followed by attenuation. The critical lengths of the nonbuffer and opening buffer structures are 2 and 6 km, respectively, whereas the critical lengths of the tunnel satisfying the control standard of the micro-pressure waves are 33 and 34 km, respectively. Although the opening buffer structure can significantly reduce the maximum pressure gradient of the initial compression waves for a long tunnel, owing to the continuous intensification of the compression wave during propagation, the effect of the opening buffer structure on the mitigation of the micro-pressure waves is significantly weakened. Engineering measures (such as shafts) should be adopted to mitigate intensification. In addition, the effects of the buffer structure on the maximum pressure gradient of compression waves are different in the portals of different tunnel lengths. Therefore, the different types of buffer structure and length factors should be combined to determine the corresponding optimal tunnel length matching relationship. 1 tab, 24 figs, 33 refs.More>

2021, 21(4): 150-162. doi: 10.19818/j.cnki.1671-1637.2021.04.011

FullText HTML (229) PDF (5504KB) (40)

Influence of surface micro shot peening on short fatigue crack behavior of CuNi₂Si alloy

YANG Bin, FENG Bo, LI Yi-fan, LIAO Zhen, ZHANG Ji-wang, XIAO Shou-ne, YANG Guang-wu, ZHU Tao

Abstract: Under tension-compression loading, the fatigue short crack replica tests of funnel-shaped round bar specimens of CuNi₂Si alloy with/without micro-shot peening were carried out, respectively. Tests were interrupted at a series of predetermined load cycles to replicate the surface of the specimen with acetate film, and then the data related to short crack initiation and propagation were obtained by using the reverse order observation method. Analysis results show that the fatigue cracks of both types of specimens initiate on the surface. The crack shows a zigzag growth under the influence of microstructure. In general, it shows a trend of slow growth in the early stage followed by a rapid growth in the later stage until the fracture of the specimen, with a critical instability crack propagation size of about 750.0 μm. With the introduction of micro shot peening technology, the crack growth process changes from the intergranular mode to the transgranular mode. The fracture morphologies of specimens with/without micro shot peening are very different. Compared to un-peening specimens, micro shot peening specimens show a larger crystal plane in the crack initiation position without obvious grain characteristics. They show smaller crack source area, larger number of fatigue stripes produced in the fatigue process, and more obvious dimple morphology in the instantaneous fracture area. After micro shot peening, the average fatigue life of the specimens increases approximately 31.5 times, and the fatigue life proportion occupied by the stage of crack initiation and slow growth increases from 60% to 80%, indicating that the significant increase in fatigue life is mainly reflected in the initiation and stable propagation stage of short cracks. The strengthening effect is mainly affected by the surface effective stress, hardness, and the number of grain boundaries, but it has little effect on the later stage of fatigue crack growth. 3 tabs, 11 figs, 30 refs.More>

2021, 21(4): 163-171. doi: 10.19818/j.cnki.1671-1637.2021.04.012

FullText HTML (101) PDF (22697KB) (19)

Dynamics modeling of asymmetrical tension and compression characteristics of gas-hydraulic and ring spring combined draft gear and impact simulation

SUN Shu-lei, LI Fu, WANG Guang-chao, TIAN Guo-ying, DENG Peng-yi, MI Cai-ying

Abstract: A dynamics model of a gas-hydraulic and ring spring combined draft gear was established to simulate the asymmetric dynamic tension and compression characteristics of the gear.Based on the characteristics of different energy-absorbing elements, a vehicle impact dynamics model program was compiled using MATLAB/Simulink software, and the impact dynamic characteristics of two vehicles and two groups of EMUs were evaluated. Research results show that the linear loading characteristics of the ring spring draft gear under tension can be simulated precisely. Moreover, the varying nonlinear characteristics of the gas-hydraulic draft gear with impact speed under compression can be simulated satisfactorily using the combined draft gear dynamics model. The asymmetric tension and compression characteristics are shown in the combined draft gear dynamics model. Under low-speed and medium/high-speed impacts, the dynamic energy absorption process and hysteresis tension-compression characteristic curve can be entirely and effectively simulated using a combined draft gear dynamics and vehicle impact model. This is caused by the deformation of the draft gear, crushing tube, anticlimbing device, and vehicle body structure in sequence. When the impact speed is 5 km·h^-1, the maximum coupler force and the maximum stroke of the combined draft gear are lower than the resistance force and stroke limit, and the loading characteristic curve of the gas-hydraulic draft gear only exhibits compression loading characteristics. When the impact speed is 20 km·h^-1, the maximum coupler force is 2 900 kN, the maximum stroke is 534 mm, and the anticlimbing device is triggered. The continuous mechanical characteristics of the gas-hydraulic draft gear, crushing tube, and anticlimbing device are exhibited in the compression-loading characteristics curve, and the structure of the vehicle body is not damaged. When the impact speed reaches 25-30 km·h^-1, the vehicle body structure is deformed, and the coupler force increases sharply. The structural safety of the vehicle body can be satisfied through the parameter selection of the fully automatic coupler and semi-permanent coupler within an impact speed of 20 km·h^-1. 2 tabs, 12 figs, 33 refs.More>

2021, 21(4): 172-182. doi: 10.19818/j.cnki.1671-1637.2021.04.013

FullText HTML (173) PDF (9168KB) (28)

Speed prediction by online map-based GCN-LSTM neural network

CHEN Hua-wei, SHAO Yi-ming, AO Gu-chang, ZHANG Hui-ling

Abstract: Online map speeds of roads were collected by calling the path-planning application programming interface of the online map to completely extract the spatio-temporal features of the road speed from road network speed and then achieve high-precision road speed prediction. The spatial features were extracted using a graph convolutional network (GCN), and the temporal features were extracted using a long short-term memory (LSTM) neural network. An online map-based GCN-LSTM neural network was established, the spatio-temporal features of the road speed were extracted, and the road speed was predicted. The performance of the online map-based GCN-LSTM neural network was assessed, and the advantages of the online map-based GCN-LSTM neural network and the substitutability of the detector-based speed prediction model were evaluated. By using the local road network as an example, the performance of the model was analyzed, and the performances of different online map-based models and similar models with different data sources were compared. Analysis results show that the mean absolute errors(MAEs) of the GCN-LSTM neural network are lower than 5, the root mean square errors (RMSEs) are lower than 6, and the mean absolute percentage errors (MAPEs) are lower than 30% in the training and testing sets. Hence, the training and testing errors are low, indicating good comprehensive performance. The MAPE of the GCN-LSTM neural network of the roads follows a Gumbel distribution, whose mean ranges between 19%±4%, and the 85% quantile ranges between 34%±5%. Hence, both indexes are low, indicating good individual performance. Among the online map-based speed prediction models, the MAE, RMSE, MAPE, mean, and 85% quantile of the MAPE fitting curve of the GCN-LSTM neural network have the lowest values. Hence, its comprehensive and individual performances are the best, and it exhibits advantages in online map-based speed prediction. Among the similar models, the MAE, RMSE, MAPE, mean, and 85% quantile of the MAPE fitting curve of the GCN-LSTM neural network have the lowest values. Hence, its comprehensive and individual performances are the best. Furthermore, the reliability of online map-based speed prediction is high, so that it can be used as a substitute for detector-based speed prediction. 4 tabs, 13 figs, 30 refs.More>

2021, 21(4): 183-196. doi: 10.19818/j.cnki.1671-1637.2021.04.014

FullText HTML (374) PDF (11679KB) (188)

Characteristics of driving speed at tunnel entrance and exit at high speed

XU Jin, ZENG Yue

Abstract: In order to clarify the operating characteristics and driving behavior at the entrance and exit of mountain tunnels and reveal the mechanism of traffic accidents at tunnel portals, three tunnels on expressways and urban express roads were selected, respectively. The spot speeds of passenger cars and trucks at the entrances and exits of these tunnels were collected. The observation sample size for a single section on the expressway is more than 500, for the urban fast road tunnel, it is more than 1 100. The variation rules and influencing factors of vehicle speed were analyzed based on the section data, and the predictive model of vehicle speed was developed. Analysis results show that drivers will slow down vehicle speed when approaching the tunnel portal. The speeds of passenger cars and trucks decrease by 12-21 and 2-10 km·h^-1, respectively. The speed decrease of truck is lower than that of passenger car. Near the tunnel portal, the speeds of passenger cars and trucks are more than 80 and 70 km·h^-1, respectively. The speed range of the expressway tunnel entrance and exit section is 75-110 km·h^-1, for the entrance section of the fast road tunnel, it is 60-88 km·h^-1. The speed at entrance or exit of the expressway tunnel is generally higher than that of the urban fast road tunnel. Drivers speed up after acclimatizing in the tunnel and accelerate when they depart the tunnel. However, they slow down when there is a sharp curve and interchange in front of the tunnel exit. The deceleration of vehicle within 100 m in front of the tunnel entrance is the largest. The decelerations of trucks and passenger cars are 0.23-0.58 and 0.47-0.70 m·s^-2, respectively. The speed data measured at the same section have strong discreteness, indicating an obvious longitudinal interference among various vehicles, which is easy to cause rear-end collision. 5 tabs, 8 figs, 30 refs.More>

2021, 21(4): 197-209. doi: 10.19818/j.cnki.1671-1637.2021.04.015

FullText HTML (318) PDF (18372KB) (96)

Subway passenger flow prediction based on optimized PSO-BP algorithm with coupled spatial-temporal characteristics

HUI Yang, WANG Yong-gang, PENG Hui, HOU Shu-qian

Abstract: To improve the accuracy of subway passenger flow prediction, by considering the Xi'an Metro Line 1 as an example, five main factors affecting subway passenger flow variations, such as festival, non-festival, time period, station, and weather, were extracted to analyze the coupled spatial-temporal characteristics of subway passenger flow. A back propagation (BP) neural network was constructed to predict the subway passenger flow. The proposed BP neural network was further optimized by using a particle swarm optimization (PSO) algorithm that introduced adaptive mutation and balanced inertia weights to form a subway passenger flow prediction system that could consider complex influence factors. Transfer stations and non-transfer stations including a first and an intermediate station were selected, the weather, festival, and non-festival factors were considered, and the BP neural network models for different time periods were compared. Then, the prediction errors of the PSO-BP neural network model were optimized. Research results show that by considering the weather, festival and non-festival factors, the mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) of the optimized PSO-BP neural network model predictions at transfer stations within the optimized time periods decrease by 40.13%, 31.46% and 23.89%, respectively, compared with the optimized PSO-BP neural network models prediction errors without the time periods, decrease by 17.50%, 17.86% and 17.32% compared with the BP neural network models prediction errors within the optimized time periods. The MAE, RMSE, and MAPE of the optimized PSO-BP neural network model predictions in the non-transfer stations within the optimized time periods decrease by 16.50%, 20.99% and 32.59%, respectively, compared with the optimized PSO-BP neural network model prediction errors without time periods, and decrease by 11.48%, 12.10% and 17.73%, respectively, compared with the BP neural network model prediction errors within the optimized time periods. The MAE, RMSE, and MAPE of the optimized PSO-BP neural network model predictions at each station within the optimized time periods decrease by 24.37%, 24.48% and 29.69%, respectively, compared with the optimized PSO-BP neural network model prediction errors without time periods, and decrease by 13.49%, 14.02% and 17.59%, respectively, compared with the BP neural network model prediction errors within the given time periods. Therefore, using the optimized PSO-BP neural network model and considering the influencing factors can improve the accuracy of subway passenger flow prediction. 8 tabs, 12 figs, 30 refs.More>

2021, 21(4): 210-222. doi: 10.19818/j.cnki.1671-1637.2021.04.016

FullText HTML (349) PDF (4815KB) (122)

Multi-task perception algorithm of autonomous driving based on temporal fusion

LIU Zhan-wen, FAN Song-hua, QI Ming-yuan, DONG Ming, WANG Pin, ZHAO Xiang-mo

Abstract: The sequential image frames were used as input to mine the temporal associated information among the continuous image frames, and a multi-task joint driving environment perception algorithm fusing the temporal information was constructed to rapidly detect the traffic participation targets and drivable area through multi-task supervision and joint optimization. ResNet50 was used as the backbone network, in which a cascaded feature fusion module was built to capture the non-local remote dependence among different image frames. The high-resolution images were processed by the convolution subsampling to accelerate the feature extraction process of different image frames, balancing the detection accuracy and speed of the algorithm. In order to eliminate the influence of spatial displacements of the objects among the image frames on the feature fusion, and considering the non-local dependence of the features of different image frames, the temporal feature fusion module was constructed to align and match the time sequences of feature maps corresponding to different image frames for forming the integrated global feature. Based on the parameter-sharing backbone network, the heat map of generating key point was exploited to detect the positions of pedestrians, vehicles and traffic signal lights on the road, and the semantic segmentation sub-network was built to provide the drivable area information for autonomous vehicles on the road. Analysis results show that the proposed algorithm takes sequential frames as input instead of single frame, which makes effective use of the temporal characteristics of the frames. In addition, its computational complexity with the cascaded feature fusion module greatly reduces to sixteenth of that without the cascaded feature fusion module through downsampling. Compared with other mainstream models, such as CornerNet and ICNet, the detection accuracy and segmentation performance of the algorithm improve by an average of 6% and 5%, respectively, and the image processing speed reaches to 12 frames per second. Therefore, the proposed algorithm has obvious advantages in the speed and accuracy of image detection and segmentation. 6 tabs, 9 figs, 31 refs.More>

2021, 21(4): 223-234. doi: 10.19818/j.cnki.1671-1637.2021.04.017

FullText HTML (274) PDF (15027KB) (131)

Optimization method of dynamic trajectory for high-speed train group based on resilience adjustment

SONG Hong-yu, SHANGGUAN Wei, SHENG Zhao, ZHANG Rui-fen

Abstract: The dynamic operation process of high-speed train groups was investigated to enhance the autonomy and intelligence of train control, and a distributed information interaction model of high-speed train groups was constructed based on the multi-agent and graph theoretic approaches. A multiobjective optimization model was formulated to optimize the energy saving and punctuality of train groups and ensure the safety and passengers' comfort. The static optimal trajectories of train groups were determined through the differential evolution algorithm modified based on the simulated annealing. On this basis, a resilience-based dynamic interval adjustment mechanism for the train groups supported by the information exchange was specifically established for the moving block system to prevent or eliminate the train delay propagation caused by the stochastic disturbances during the operation. Moreover, an online cooperative optimization algorithm was developed to achieve the dynamic adjustment of the train group trajectories. Finally, simulations were performed based on the actual field data of the Wuhan-Guangzhou High-Speed Railway. Research results show that the proposed online cooperative optimization algorithm can effectively improve the optimal solution searching ability, and avoid excessively frequent updates of the Pareto optimal set. The average algorithm trigger times under different disturbance scenarios decreases by 36.7%. In typical disturbance scenarios, the optimized dynamic adjustment approach decreases the delay degree of the disturbed train from 6.2% to 0, and guarantees the safe and smooth operation of the train group. The optimized approach can save the energy consumption by up to 4.8% compared with the immediate delay recovery approach. Even with more significant disturbance scenarios, the delay degree of the disturbed train decreases from 13.1% to 1.4%, and the global time deviation decreases to 0 with an energy-saving rate of 1.8%. The proposed method can solve the problem that the static trajectory planning is unable to fully adapt to the change in the external dynamic environment, and effectively and timely restore the train operation despite complex disturbances. 7 tabs, 24 figs, 31 refs.More>

2021, 21(4): 235-250. doi: 10.19818/j.cnki.1671-1637.2021.04.018

FullText HTML (191) PDF (12068KB) (44)

Rapid check method for truck loaded with fresh agricultural products on expressway

YAN Sheng-yu, XIAO Run-mou, FANG Yan-ming, WANG Yi-meng

Abstract: To improve the traffic efficiency of trucks loaded with fresh agricultural products (FAPs) and reduce the workload of inspectors at toll stations, a rapid check model for trucks loaded with FAPs based on a laser ranging system was proposed. The characteristics of main types of trucks loaded with FAPs was analysed, and common types of trucks loaded with FAPs were determined by traveling route induction. The laser ranging system of carriage outline was introduced and the cargo density database was established as the basis of judgment. According to the required loading limits of tolling for free, and considering the probability distribution of cargo densities, the upper and lower thresholds of the cargo density were determined, and a cargo density judging model for unreasonable loading was developed for the case of an 8×4 truck. Based on the features of space between truck axles, the moment balance principal (MBP) was used to deduce the theoretical weight of the assembled axles, which were close to the front of the carriage. The assembled axles judging model was then developed based on the MBP. The feasibilities of the cargo density judging model and assembled-axles judging model were verified by the measuring data of toll stations, the computational errors of the two models were analyzed, and the advantages of the rapid check model were compared with those of the manual model. Analysis results show that there are five main types of trucks in FAP transport, The deviation rates of volume and density measured by the cargo density judging model are ±3.75% and ±4.94%, respectively, which is considerably less than 50% of the tolerance of a reasonable range or warning range in the judging logic. The dependence of the cargo density judging model based on the laser ranging system is less than that of the assembled axles judging model. The efficiency of the rapid check model based on the two models is 5.97 times higher than that of the manual inspection, which can effectively improve the efficiency of the lane. The inspection rate of all the investigated trucks is 16.62%, which can reduce the inspection frequency and economize labor and employment. Although the average time of the added selection and input parameter process is 0.41 min, it is only 15.70% of the time of the manual inspection. 4 tabs, 3 figs, 30 refs.More>

2021, 21(4): 251-258. doi: 10.19818/j.cnki.1671-1637.2021.04.019

FullText HTML (191) PDF (2464KB) (29)

Effect of fog weather warning system under cooperative vehicle infrastructure on vehicle operating eco-characteristics

WU Yi-ping, LI Hai-jian, ZHAO Xiao-hua, XING Guan-yang, CHEN Yu-fei, FU Qiang

Abstract: To explore the effectiveness of cooperative vehicle infrastructure (CVI) technology on vehicle operating eco-characteristics, a fog weather warning system under CVI condition was built based on the driving simulation experiment platform. The fuel consumption and emission features of driving vehicles under the heavy fog weather environment were tested. Four testing scenes, including blank control group, group of warning by dynamic message sign (DMS), group of warning by human machine interface (HMI), and group of warning by DMS+HMI, were designed. A driving simulator experiment was carried out by recruiting 43 drivers. The effects of different warning methods on vehicle operating eco-characteristics were obtained by comparing the differences in vehicle fuel consumption and emissions as a whole and on key sections of the road. Analysis results show that compared with the blank group, the three types of fog weather warning systems under CVI can significantly reduce the overall fuel consumption and emissions of vehicles. However, the effects of the different warning methods are not significantly different. The road scene is divided into pre-warning, warning, gradient and fog zones, and the three warning systems can effectively reduce vehicle fuel consumption and emissions in the warning and gradient zones. The HMI takes effect as the warning message is issued, and the DMS is effective before the vehicle enters the warning zone. The DMS+HMI has the most significant effect in the warning zone, but it can not effectively reduce the vehicle fuel consumption and emissions after the vehicle enters the fog zone. Therefore, although fog weather warning system under CVI can improve the overall eco-characteristics of vehicle operation, it can not effectively guarantee energy savings and emission reduction effect of different sections in the fog-affected area by only increasing the warning intensity or changing the warning mode. The matching relationship among different warning modes, warning information trigger points and timing, driver characteristics and other factors should be comprehensively considered in a reasonable setting of the fog weather warning system under CVI. 4 tabs, 8 figs, 30 refs.More>

2021, 21(4): 259-268. doi: 10.19818/j.cnki.1671-1637.2021.04.020

FullText HTML (191) PDF (10047KB) (78)

User Center

Journal Info

Collection of Source JournalsMore>

2021 Vol. 21, No. 4

Search Type

Search Journal

LinksMore>