西南交通大学学报

Simplified Calculation of Temperature in Concrete Slabs of Ballastless Track and Influence of Extreme Weather

LIU Xueyi, LI Jiali, KANG Weixin, LIU Xiaokai, YANG Rongshan

2017, 30(6): 1037-1045,1060. doi: 10.3969/j.issn.0258-2724.2017.06.001

[Abstract](558)

Abstract:
In order to ascertain the temperature distribution characteristics of a ballastless track, heat transfer mechanisms between the concrete slab and the environment were investigated, and a field temperature test of the ballastless track was conducted. The spectrum analysis method was used to study the relationship between meteorological data and the measured concrete slab temperature. Further, based on thermodynamic principles and experimentation, concrete slab thermal parameters were calculated, and a simplified concrete slab temperature calculation method was established. The predicted and measured results were then compared, the influence of solar radiation, wind, and air temperature on track temperature were studied, and concrete slab temperature distribution characteristics under extreme weather were also analysed. The results show that this calculation method has good accuracy (0.4-3.0℃) and applicability in calculating concrete slab temperature. The concrete slab thermal diffusivity was determined to be 0.004 3 m2/h. Mean air temperature and solar radiation play a leading role in overall concrete slab temperature, and the air temperature amplitude greatly influences the concrete slab temperature gradient. When the wind scale is greater than 4, the concrete slab surface temperature is approximate to the equivalent air temperature. During extreme weather, the temperature at the lower concrete slab surface can reach 50℃, and the temperature gradient can reach 110℃/m.

Experiment and DEM Analysis of Lateral Resistance of Glued Ballast

XIAO Hong, LING Xing

2017, 30(6): 1046-1054. doi: 10.3969/j.issn.0258-2724.2017.06.002

[Abstract](532)

Abstract:
Lateral resistance data were obtained through a field experiment in a high-speed railway to reveal the working mechanism of the lateral resistance of glued ballast. The change characteristics of the lateral resistance of the glued ballast were analysed. Moreover, a three-dimensional discrete element model was established by PFC3D. The contact force and stress of the glued ballast were also statistically analysed. The results show that the lateral resistance of the glued ballast is significantly improved, and its value is 4.6 times of the standard value of the ballast bed. The ballast can maintain a good force state under its high-lateral resistance state. The maximum value of the contact force between the sleeper and the ballast is 1.2 KN, and the average value is 112.48 N. When the whole section is bonded, a depth range of 26 cm must be ensured to be fully and uniformly bonded to ensure the bonding effect. Compared with the ballast bed, the sharing ratio of the lateral resistance significantly varies in different glued ballast positions. Furthermore, the lateral resistance of the glued ballast is mainly shared by the sleeper side, followed by the sleeper bottom and the shoulder with a sharing ratio of approximately 63%, 24%, and 13%, respectively.

Analysis of Mechanical Characteristics of Degradation Railway Ballast by Direct Shear Test

JING Guoqing, HUANG Hongmei, CHANG Jinxiu, QIANG Weile

2017, 30(6): 1055-1060. doi: 10.3969/j.issn.0258-2724.2017.06.003

[Abstract](597)

Abstract:
In order to study the mechanical characteristics and reuse of a washed recycled ballast, a Los Angeles abrasion tester and strain-controlled direct shear apparatus were used in experiments to investigate the stress-strain characteristics of recycled ballast with different proportions of fresh ballast under different normal stresses (50, 100, and 200 kPa). Based on the experimental results, the mechanical features of the ballast were discussed. The results show that ballast strength increased with the proportion of fresh ballast. The strength of mixtures with 50% and 100% fresh ballast were 143.4 and 176.7 kPa, respectively, under normal stress of 100 kPa; an increase of 24.0% and 52.9%, respectively, compared to pure recycled ballast. The shear dilatation of the mixture also increased with the proportion of fresh ballast. For 50% and 100% fresh ballast samples, the values are 9.842 mm and 7.969 mm, respectively, under normal stress of 100 kPa; a decrease of 25.8% and 39.9%, respectively, compared to pure recycled ballast. There was a small difference in mechanical characteristics between washed recycled ballast and fresh ballast, and the washed recycled ballast can be reused with fresh ballast under certain conditions.

Fatigue Life Assessment of Full-Penetration Cruciform Welded Joints under Tensile Shear Cyclic Stress

WEI Xing, JIE Zhiyu

2017, 30(6): 1061-1067. doi: 10.3969/j.issn.0258-2724.2017.06.004

[Abstract](535)

Abstract:
To investigate the fatigue performance and fatigue life assessment method of full-penetration welded joints in tensile shear cyclic stress, fatigue tests and finite element analyses were carried out. With different inclined angles of weld lines of 0°, 15°, 30°, and 45°, full-penetration cruciform welded joints were designed at different tensile shear stress ratios. Using the ANSYS finite element program, three-dimensional (3D) finite element models of cruciform welded joints were developed and the stress distributions and stress concentration factors were investigated. Four groups of fatigue tests (12 specimens) were conducted under proportional loading over the same stress range as that on the end of the steel plates. Based on the nominal stresses, the fatigue lives of full penetration cruciform welded joints were evaluated by using effective stress, interaction equations, and bi-parametrical critical plane approaches. The accuracy and reliability of the assessment methods were checked against the experimental results of the fatigue testing. The research results showed that the fatigue life calculated using the approach of the International Institute of Welding (ⅡW) and Eurocode is longer than the test result, whereas the fatigue life based on the principal-tensile-stress method is shorter than the test result. Furthermore, fatigue life given by the effective stress and bi-parametrical critical plane approaches are in good agreement with the test results.

Study of Train Running Performance under Wave Load for Cross-Sea Bridge

FANG Chen, LI Yongle, XIANG Huoyue

2017, 30(6): 1068-1074. doi: 10.3969/j.issn.0258-2724.2017.06.005

[Abstract](460)

Abstract:
The Pingtan Strait Bridge is in a complex and harsh marine environment. The wave loads could affect traffic safety and comfortability.Based on the numerical simulation for coupling vibration of the rail vehicle-bridge system,our software-BANSYS (bridge analysis system) was adopted for analysing the dynamic responses of vehicles and the bridge under extreme wave loads.To estimate the return periods of the wave load, velocities, water depths, stiffness of bridge foundations, and the responses of vehicles and the bridge were analysed.The results show that wave loads have a significant impact on the responses of the vehicle-bridge system; when the return period of the wave load is 50 years, the lateral displacement of the midspan exceeds the standard limit. The wave loads play a major role in the train running performance; with large wave loads, the dynamic response of the vehicle-bridge system is more sensitive to vehicle speed than with small wave loads. The scheme of low pile cap can effectively reduce the dynamic response of the bridge and vehicles under wave loads. The different frequently-used concrete strengths of the bridge foundations have little influence on traffic safety and comfortability, and the relative change in the maximum acceleration of vehicles is up to 3%.

Analysis of Dynamic Damage Constitutive Model for Concrete Exposed to High Temperature

WANG Fei, QIAN Yongjiu, XU Jinyu, LI Zhiwu

2017, 30(6): 1075-1081. doi: 10.3969/j.issn.0258-2724.2017.06.006

[Abstract](565)

Abstract:
To fill a gap in the field of dynamic constitutive model research of concrete, a dynamic damage constitutive model for concrete was established based on a nonlinear viscoelastic constitutive model and damage mechanics, while considering the effect of damage on the mechanical properties of the material. The proposed dynamic damage constitutive model contained nine parameters such as material parameters and modulus of elasticity. High temperature is considered to directly influence the material parameters of the model, and these model parameters were obtained for various temperatures by fitting the experimental stress-strain curves. The comparison of stress-strain curves between the model predictions and experimental results shows that the error of peak stress, peak strain, and modulus of elasticity is less than 5%, 9%, and 7%, respectively, and the fitting similarities are greater than 0.93. Consequently, the model can be used to simulate the mechanical behavior of concrete after exposure to high temperatures.

Refined Finite Element Modeling Methodof Spatial Prestressed Steel Beam

WU Yanbin, HUANG Fanglin

2017, 30(6): 1082-1087. doi: 10.3969/j.issn.0258-2724.2017.06.007

[Abstract](582)

Abstract:
In order to establish a continuous smooth model of spatial prestressed steel beam include both horizontal and vertical curvature, to consider the mechanical effects of spatial prestressed steel beam in finite element analysis, the refined finite element modeling method of spatial prestressed steel beam based on intersecting line was proposed, so the connection mode of the nodes between the spatial prestressed steel beam, duct and concrete girder could be ascertained, and the corresponding modeling procedure was given. The continuous smooth geometric model of spatial prestressed steel beam was established by the intersecting auxiliary bodies. The nodes corresponding relationship between steel beam, duct and concrete girder were achieved by establishing the solid model of duct and performing Boolean operations with main girder. Take the finite element modeling and analysis of the steel-concrete joint section of the Hongshui River bridge in Guizhou as an example, the results show that the maximum local stress of girder decreased by 26.3%, and the stress concentration is improved obviously compared with the traditional constraint equation method, the effective prestress distribution of steel beam is smooth, non-oscillating, the proposed method is more in line with the actual situation.

Model Test Study on Long-Span Railway Concrete Arch Bridge with Rigid Skeleton

ZHANG Shuangyang, ZHAO Renda, JIA Yi, WANG Yongbao, XIE Haiqing

2017, 30(6): 1088-1096. doi: 10.3969/j.issn.0258-2724.2017.06.008

[Abstract](572)

Abstract:
The Beipanjiang bridge is a 445 m concrete arch reinforced rigid skeleton bridge on the high-speed railway from Shanghai to Kunming. The bridge deck is designed to comply with design rail speeds of 350 km/h and subsequently requires strictly controlled stresses and deformation for this ballast-less track. To verify the safety of the bridge during construction and in service and to rationalise the use of outsourced concrete construction methods, a 1:7.5 scaled model based on the prototype bridge was constructed. The loading method and arrangement of control sections and measuring points are discussed. Comparing the test results and computed results provides a comprehensive understanding of the mechanical behavior of the bridge. The results indicate that the maximum stress of the steel tube is 250 MPa, located at the upper chord of 1/2 cross section and the maximum stress of the C60 outsourced concrete is 17 MPa, located at the side-box bottom of the half cross section, whilst the maximum displacement of the arch ring is 68 mm, also positioned at the half cross section. Considering the stress equivalent and geometric similarity principles, the comparative results indicate that the stress and displacement of the model bridge align with the design parameters of the prototype. Model test results prove that the stress and deformation of the prototype bridge can meet the requirements during construction.

Structural Reliability Design of Non-Backfilling Arch Open Tunnel Structure under Rock-fall Impaction

WANG Yusuo, XU Ming, WANG Tao, WANG Zhilong, HE Junnan, ZHOU Liang

2017, 30(6): 1097-1103. doi: 10.3969/j.issn.0258-2724.2017.06.009

[Abstract](562)

Abstract:
A structural reliability design method for a non-backfilling arch open tunnel structure under rockfall impaction was investigated to improve the reliability of structural design. Firstly, in view of the characteristics of the failure modes of a non-backfilling arch open tunnel under rockfall impaction, a four-fixed square reinforced concrete slab was simplified to a rigid perfectly-plastic slab, and the assumed failure of the structure's ultimate load or structural resistance was calculated as a statics problem using plastic limit theory. Secondly, numerical modelling was performed using a particle flow code method, and rockfall impaction load was expressed as a power function of the gravitational potential energy of rockfall through the regression analysis of modelling results. Thirdly, the limit state equation was established by combining structural resistance and the rockfall impaction load expression. Then, the structural reliability index was obtained using a MATLAB software programming method, and structure reliability design and optimization were performed by comparing the calculated structural reliability index with the target reliability index. Finally, using the established method, the structural reliability indexes of the double-line arch open tunnel of a passenger transport line were analysed. The structural reliability index of the original structure design is up to 4.2 for a height of 5-15 m and a rockfall weight of 1-2 kN. Moreover, the structural reliability index is up to 5.4 if the concrete grade is adjusted from the original C35 to C40 and the reinforcing steel bar grade is changed from HRB335 to HRB500, for a height of 5-20 m and a rockfall weight of 1-2 kN.

Effects of Material Damage and Structural Characteristics on Tunnel Shield During Loading and Unloading

LU Daiyue, XU Guowen, WANG Shimin

2017, 30(6): 1104-1112. doi: 10.3969/j.issn.0258-2724.2017.06.010

[Abstract](550)

Abstract:
To study the effects of material damage and structural characteristics in a tunnel shield during loading and unloading, we conducted an investigation using a plastic-damage constitutive model for the concrete material, nonlinear contact theory, Python's secondary development, and a unidirectional compression spring around the shield circle, which represented the ground reaction force. A Two-dimensional (2D) discontinuous segment-joint model along with a three-dimensional (3D) precision discontinuous segment-joint model were built to simulate the tunnel shield's structural distortion and internal forces when exposed to varying material damage conditions during both loading and unloading. These conditions were explored using three different soil-condition scenarios. The influence of bolt abnormality on the structure is discussed, with graphs showing the corresponding soil radial displacement, the joint stretching value, material stress, and the damage factor for both the loading and unloading condition. Graphs for the joint stretching value and bolt stress levels are presented for two kinds of abnormalities:pre-tightening force loss and corrosion depth. The results show that the deformation index for the macro-structure given three kinds of soil conditions are 510, 340, and 170 kPa, which are considered safe values for loading and unloading. Regarding material stress level to avoid extensive damage, the safe values for loading and unloading are 340, 170, and 170 kPa. The influence of bolt preload loss on the structure is primarily controlled by the opening of the dome joint. The influence of bolt corrosion on the structure is primarily controlled by the stress level of the bolt. The safety limit for the corrosion depth is 6 mm.

Research of Shaking Table Model Tests on Immersed Tunnels under Different Conditions

CHENG Xinjun, JING Liping, CUI Jie, LI Yongqiang, DONG Rui

2017, 30(6): 1113-1120. doi: 10.3969/j.issn.0258-2724.2017.06.011

[Abstract](511)

Abstract:
To study the seismic response and anti-seismic behaviour of immersed tunnels at different sites, shaking table tests were conducted in three sets. The reduced-scale ratio of the tests was 1:30. The site conditions adopted in the tests were dry sand, saturated sand, and saturated sand under an overlying water layer. The experimental results show that when the peak ground acceleration (PGA) of the input excitation is 0.1g, the acceleration amplification factors of dry sand, saturated sand, and saturated sand under an overlying water layer exhibit a magnifying trend. No soil liquefaction occurred in the saturated sand sites. When the input PGAs are 0.2g and 0.4g, the acceleration of the soil deposits of the dry sand site continues to increase, whereas the acceleration of the surface points in the saturated sand sites decreases owing to sand liquefaction. When the input PGA is 0.4g, the maximum reduced acceleration percentages of the middle tunnel elements in the three sets of tests are11.1%, 7.6%, and 9.4%, respectively, compared with those of the end tunnel elements. The flexible joints have a shock absorption effect during the shaking process. The development trends of pore water pressure of the soil deposits correlate well for the saturated sand and saturated sand under an overlying water layer. In addition, the results of the three sets of tests indicate that the nodes between the middle wall and the base slab are the weak positions of the tunnel model.

Comparison of 2D and 3D Dynamic Responses of Saturated Soft Soil due to Metro Train Load

HUANG Qiang, YE Bin, HUANG Hongwei, ZHANG Dongmei, ZHANG Feng

2017, 30(6): 1121-1129. doi: 10.3969/j.issn.0258-2724.2017.06.012

[Abstract](519)

Abstract:
As for the common practice that the two-dimensional(2D) model is approximately employed to the three-dimensional(3D) model in numerical calculation of geotechnical engineering, 2D and 3D dynamic responses induced by the metro train vibration were analyzed. Train vibration load on the track bed was determined firstly according to the rail-fastener-tunnel-subgrade longitudinal model, then 2D and 3D dynamic responses of the tunnel and the saturated soft ground were simulated based on the cyclic mobility model and the fully coupled soil-water theory when the metro train moved at an average speed. The results show that:dynamic responses of the ground surface acceleration, displacement and excess pore water pressure (EPWP) around the tunnel in 2D and 3D models are similar but significant differences exist in the quantitative manner.The maximum acceleration and displacement ratios between 2D and 3D models can be as much as ninefold and sixfold, with a maximum acceleration level discrepancy of 15 dB; the ratio of EPWP around the tunnel ranges from 1.5 to 3.5, and the accumulated EPWP in 2D and 3D models reach 4.36 kPa and 1.69 kPa, respectively, the accumulation of EPWP is obvious at the locations of 45° and 135° around the tunnel. Train vibration load, longitudinal soil vibration and consolidation rate are the three reasons accounting for the response discrepancy between 2D and 3D models.

Buckling Analysis for Tapered Pile Considering Self-Weight of Pile

LI Chuanxun, LIU Xiaozhao, WU Wenbing

2017, 30(6): 1130-1138. doi: 10.3969/j.issn.0258-2724.2017.06.013

[Abstract](519)

Abstract:
To study the effect of pile shape on the buckling behaviour of a tapered pile, the first-type stability of the tapered pile was analyzed based on the cusp catastrophe model of the pile-soil system with considering the self-weight of the pile. By determining the potential function and the equations of the bifurcation sets of the pile-soil system, the cusp catastrophe model was established, and the instability condition of the pile-soil system was derived based on the cusp catastrophe theory and the principles of the energy method. Meanwhile, the buckling load of the tapered pile was determined by the Rayleigh-Ritz method, and the reliability of the results from the cusp catastrophe theory was verified. Finally, the influences of the self-weight of the tapered pile and the variation ratio of pile diameters on the buckling load were analyzed. The result shows that when h/L increases from 0 to 0.4, the ratio of the buckling load reduction of the tapered pile with considering the self-weight of the pile to the buckling load without considering the self-weight of the pile decreases to 42.82%, and the ratio of the buckling load reduction for an equal-diameter pile with considering the self-weight to the buckling load without considering the self-weight decreases to 75.27%. Under the identical conditions, the greater the tapering angle of the tapered pile, the better the stability. When the dimensionless length of the pile tends to be zero, the calculation length of a tapered pile with a free end and a fixed end tend to be zero and that of a tapered pile with a free end and a hinged end tend to be infinite.

Mapping Relationship between Rail Deflection of Slab Track and Subgrade Settlement

GUO Yu, GAO Jianmin, SUN Yu, ZHAI Wanming

2017, 30(6): 1139-1147,1215. doi: 10.3969/j.issn.0258-2724.2017.06.014

[Abstract](561)

Abstract:
In order to investigate the effect of subgrade settlement on the track geometry of the slab track, a calculation method for the mapping relationship between rail deflection and subgrade settlement was proposed, aiming at both longitudinal connected track and unit slab track of high-speed railways. The method is based on the theory of composite beams on tensionless Winkler foundation, which takes into account the potential voids between the bottom slab and the subgrade. The characteristics of the rail deflections of the aforementioned tracks, induced by different settlement profiles, were investigated and compared. The results show that in terms of the adaptability to the subgrade settlement, the longitudinal connected track is inferior to the unit slab track, while the large-scale unsupported area is more likely to be induced between the longitudinal connected track and subgrade. The difference in rail deflection between the two tracks is significantly aggravated by the amplitude of the subgrade settlement. The rail deflection is also sensitive to the settlement wavelength. When the wavelength is extended to 20 m, the number of voids under the track are significantly reduced and the rail deflection is basically in accordance with the subgrade. The influence of angular-type differential settlement on the deflection is relatively low, while the faulting-type settlement always results in severe rail deflection and voids. For instance, the maximum depth of void reaches approximately 8 mm when the faulting is 10 mm. The induced track irregularity and voids may pose threats to the train's operation quality and the fatigue behaviour of the track structure.

Flexural Performance of Composite Bamboo-Concrete Beams

WEI Yang, ZHOU Mengqian, JI Xuewei, DUAN Maojun, LI Guofen

2017, 30(6): 1148-1155. doi: 10.3969/j.issn.0258-2724.2017.06.015

[Abstract](545)

Abstract:
Three bamboo-concrete composite beams and one companion bamboo beam were tested to investigate the composite efficiency and the effects of shear connectors on bamboo-concrete composite beams. The shear connectors of these composite beams consisted of dowels, notches or notch-dowels. The results indicate that the composite beams connected with dowels fail owing to the tensile fracture of the bottom-most bamboo fiber. Furthermore, the composite beams connected with notches or notch-dowels fail owing to the shear connectors. The longitudinal shear failure occurred in the bamboo along the horizontal line located at the bottom of the notches. Compared to the companion bamboo beam, the load corresponding to the displacement of the 1/250 span of the bamboo-concrete composite beam increases by 3.18 to 3.74 times and the sectional stiffness is significantly enhanced. The effective sectional stiffness of these composite beams reaches up to 70% of the sectional stiffness of the full composite beams. For the composite beams connected with notches or notch-dowels, the interfacial slip is smaller than 1.2 mm, the load-deflection relationship is linear, and the connectors exhibit the performance of a rigid connection.

Seismic Fragility Analysis of Nepalese Brick-Timber Heritage Structures under Near-Fault Pulse-Like Ground Motions

PAN Yi, WANG Xiaoyue, XU Hu, XIE Dan

2017, 30(6): 1156-1163. doi: 10.3969/j.issn.0258-2724.2017.06.016

[Abstract](518)

Abstract:
To study the influence of near-fault pulse-like ground motions on the seismic performance of masonry-timber heritage structures in Nepal, a two-storey Nepalese masonry-timber structure was selected as a prototype according to the typical seismic damage mode in the Gorkha earthquake. A numerical simulation was established by means of the equivalent-frame model. The peak ground accelerations and inter-storey drift were used as indexes for ground motion intensity and the structural damage, respectively; subsequently, the IDA (incremental dynamic analysis) method was used to determine the ultimate states. Additionally, the seismic fragility curves of the structure subjected to pulse-like and non-pulse-like ground motions were compared. The results indicate that the inter-storey drift corresponding to the immediate occupancy state, life safety state, and collapse prevention state are 0.06%, 0.32%, and 0.44%, respectively. Although it is prone to the immediate occupancy state of the structures according to the damage probability curves, the probability of reaching the life safety state or collapse prevention state is relatively low.However, as the limit of life safety state is reached or exceeded, a slight PGA increment could lead to structural collapse. Further, with respect to the case experiencing non-pulse-like ground motions, the damage probabilities for immediate occupancy state, life safety state, and collapse prevention are to 18.9%, 14.3%, and 12.6% higher than the pulse-like ground motion case, respectively. In addition, due to the sensitivity of Nepalese masonry-timber heritage structures, the effect of pulse-like ground motions should be considered in practical reconstruction and retrofitting projects.

Study of Seismic Performance and Displacement Ductility of Confined Masonry Structures

SU Qiwang, XU Ziyi, SONG Jirong, LI Yongzhi, WU Hao

2017, 30(6): 1164-1172. doi: 10.3969/j.issn.0258-2724.2017.06.017

[Abstract](536)

Abstract:
In order to research seismic behaviours of masonry structures confined by tie-columns and beams, a 1:2 pseudo-static model test of a two-storied brick masonry building with confined elements was conducted. The mechanical characteristics, the strain development of tensioned bars in the constructional columns, and the structural skeleton curve were studied in the test. The results of the model test were compared using a numerical analysis technique, and the rationality of the parameter settings was verified. In addition, a quantitative analysis of the relationship between the change in the displacement ductility coefficient and the increment of many factors of influence including the number of stories, floor height, brick walls, and constructional columns on the floors, and the quantitative expression of the function are presented. The analysis results show that tie-columns, beams, and walls can undergo the same deformation mechanism, constraint and limit deformation, and the development of cracks. The damage to masonry structures can exhibit ductile characteristics. Increasing the number of walls can improve the lateral bearing capacity of the structures, but it cannot increase the displacement ductility, whereas increasing the number of tie-columns can significantly increase the displacement ductility. The displacement ductility coefficient can be increased by three times the increase in the number of constructional columns.

Axially Loaded of Concrete-Encased Concrete Filled Steel Tubular Stub Columns after Fire

XIANG Kai, PAN Yanchong, ZHAO Bi, WANG Guohui

2017, 30(6): 1173-1181. doi: 10.3969/j.issn.0258-2724.2017.06.018

[Abstract](476)

Abstract:
In order to investigate the failure mechanism and structural performance of axially loaded concrete-encased concrete filled steel tubular (CFST) stub columns after fire exposure, the failure mode, axial compression ultimate bearing capacity and axial stiffness of concrete-encased CFST stub columns were analyzed based on experimental study and finite element model (FEM) simulation. A total of 8 concrete-encased CFST stub columns specimens with different cross-section types, steel tube dimension and slenderness ratios were tested after fire exposure, and the nonlinear FEM simulation was carried out by ABAQUS software. The results show that compressive failure modes are observed for concrete-encased CFST stub columns, which is similar to reinforced concrete stub columns. Compared to a specimen at ambient temperature, the drop of axial compression ultimate bearing capacity of concrete-encased CFST stub columns is about 27.4%-38.2%, and the drop of stiffness is about 61.4%-69.0%. When the fire exposure time and the size of the specimens are the same, and the thickness of the reinforced concrete cover improve to 1.53 times, the rate of decline of axial compression ultimate bearing capacity and axial stiffness for circular specimens are 8.4% and 1.3%, respectively, and for the square specimens, the rate of decline of axial compression ultimate bearing capacity and axial stiffness are 5.8% and 4.1%, respectively. Compared with the FEM simulation results, the mean and standard deviation of simplified method of calculation error are 4.9% and 5.3 for axial compression ultimate bearing capacity, the mean and standard deviation of axial stiffness calculation error are 13.8% and 9.8.

Real-time Detection of Defects in Train Bearings Based on Analysis of Signal Characteristics in Time-Frequency Domains

YONG J. Yuan, HUANG Xiaoxia, TANG Yuantian

2017, 30(6): 1182-1187. doi: 10.3969/j.issn.0258-2724.2017.06.019

[Abstract](483)

Abstract:
To realize the real-time monitoring and quick diagnosis of defects in rolling trainbearings, an approach is proposed based on the time and frequency analysis of signal characteristics, especially the combination of time-domain characteristics analysis and resonance demodulation based on wavelet transforms. A real-time detection system is set up for detecting train bearing defects including a sensor data acquisition module, a sound diagnostic module, data storage and an output module. The following tests for bearings were conducted for comparison:normal bearings, bearings with inner-ring defects, and bearings with inner-ring and roller defects. The results showed that the analysis of time-domain characteristics can be used to determine bearings were faulty prior to operation. The resonance demodulation based on wavelet transform extracted the inner-ring defect frequency and the roller defect frequency of 35 Hz and 23 Hz, respectively. The stability and reliability of the real-time detection system are verified.

Friction Performance of Brake Disks and Blocks for High-Speed EMU Trains in Cold, Rainy, and Snowy Weather

QIAN Kuncai, WU Shezhang, QIAO Qingfeng, CHEN Xin, WANG Zehua

2017, 30(6): 1188-1192. doi: 10.3969/j.issn.0258-2724.2017.06.020

[Abstract](572)

Abstract:
In order to improve the safety of high-speed train operation in cold weather, the braking performance of a high-speed EMU train in temperature range from -15℃ to 20℃ in rain and snowy conditions was simulated and the causes of decline in braking force and scratching of the brake disk surface were investigated. The results showed that the average friction coefficient increased with increasing braking force. The effect of external hard particles on the friction coefficient varied with the braking speed. When the speed was less than 160 km/h, the friction coefficient increased with increasing speed. When the speed was more than 160 km/h, the external hard particles did not have an obvious effect on the friction coefficient. It was also found that the decline in transmission efficiency of the clamp in severely low temperatures and wet conditions was due to the clamp being covered with snow and becoming difficult to reset, which in turn reduced the clearance of the brake disk and brake blocks. In such a situation, external hard particles or contact between the brake block and disc would result in scratches on the brake disk surface. The research showed that braking for 1 min after every 7 min interval could remove the ice on the brake disk surface and prevent the clamp from being frozen in ice.

Critical Station Identification Based on Passenger Propagation in Urban Mass Transit Network

HU Yingyue, CHEN Feng, CHEN Peiwen, WANG Zijia

2017, 30(6): 1193-1200,1215. doi: 10.3969/j.issn.0258-2724.2017.06.021

[Abstract](526)

Abstract:
In order to reflect the real-world operation situation of an urban mass transit network, on the basis of complex network theory, the influence of passenger flow was considered to improve the recognition accuracy of the key stations in the network. By analyzing the sources of passenger flow in each section of the urban mass transit network, passenger propagation models were proposed according to the functional characteristics of passenger transport in ordinary stations and transfer stations. The model parameters were determined statistically by network flow assignment according to historical smart card data. Then, in combination with the degree and betweenness concepts in a complex network, four indexes were proposed to identify the critical stations. Taking the subway network of a certain city as an example, the dynamic passenger flow in the whole network was demonstrated and the critical stations were revealed by using smart card data on the morning peak hours of a certain workday. Research results show that the critical lines are Line 1 and Line 10. Passengers become stranded at the South Railway Station, Xi'erqi Station, and Tiantongyuan Station. The stations exposed to huge passenger flows are more vulnerable to the impacts of large passenger flow. The developed passenger propagation model can display the levels and variations in each section and station where passengers become stranded dynamically. The model can also identify the critical stations by considering the indexes of real passenger flow volume, transport capacity, and network structure. This will provide a theoretical reference for security management of urban mass transit with more efficiency.

Biochar Incorporation into Bioretention for Enhanced Ammonium Removal and Runoff Retention

TIAN Jing, LIU Dan

2017, 30(6): 1201-1207. doi: 10.3969/j.issn.0258-2724.2017.06.022

[Abstract](472)

Abstract:
Incorporating biochar into bioretention filter media was proposed to enhance urban stormwater treatment. Pinewood biochar pyrolysed at 550℃ was used to investigate its impact on the physicochemical properties of bioretention filter media and their ammonium sorption capacities in artificial stormwater. Water retention curves of the control and 4% biochar-amended filter media were measured at -15 300 to 0 cm H2O suction pressure. Hydraulic conductivities of two filter media were measured at field pilot-scale bioretention using a tension disc infiltrometer and analysed with DISC software. The results indicate that biochar amendment increased the pH of the acid filter medium, decreased bulk density by 7%, and increased total porosity and cation exchange capacity by 28% and 33%, respectively. Biochar incorporation increased ammonium removal of the control filter medium from -26%-28% to 50%-90%. The Freundlich model fitted the ammonium sorption isotherms of the three materials well. Biochar increased available water content of the control filter medium from 0.016 to 0.063 cm3/m-3 and saturated hydraulic conductivity by 1.5 times. The Van Genuchten model fitted the water retention curves of the control and biochar-amended filter media well. Incorporating biochar into bioretention filter media can enhance nitrogen removal from the perspective of water quality and quantity and benefit urban rainwater flood control.

New Contact Force Measuring Method for Overhead Catenary Systems Based on Strain

TAN Mengying, ZOU Dong, LI Ruiping, LI Yan, ZHOU Ning, MEI Guiming, ZHANG Weihua, HUANG Guanhua

2017, 30(6): 1208-1215. doi: 10.3969/j.issn.0258-2724.2017.06.023

[Abstract](586)

Abstract:
Pantograph dynamic contact force is the key factor for assessing the current collection quality of a pantograph, operation conditions of the catenary, abrasion on the pantograph slide, and the localized defect detection of the catenary. Taking into account the drawbacks of currently used methods, a new contact force measuring method based on optical fibre strain sensors is proposed. First, the pantograph head was simplified to a freely supported beam, the strain responses under the effect of a concentrated force acting on the pantograph head were analysed, and a new model for calculating the contact force was developed. Next, the linear relationship between the strain responses and the acting force was proved via static tests, and the equivalent stiffness and equivalent strip length were calibrated via static tests. The veracity of the proposed measuring method was proved via dynamic tests. Furthermore, the reliability of the method at high frequencies was verified via simulation. Finally, the contact force measured before and after inertial force correction were compared to the corresponding actual values. The results indicate that the effect of the inertial force on the contact force increases with increasing frequency. The measurement errors of the minimum and maximum values of the contact force, without considering the inertial force, are 26.57% and 11.13%, respectively, at 10 Hz. These large error values indicate that the inertial force cannot be ignored. After inertial force correction, the measurement errors of the minimum and maximum values of the contact force reduced to 3.12% and 1.54%, respectively. Thus, the measured values of the contact force are in good agreement with the corresponding actual values.

Delay-Constrained Mobile Energy Charging in Wireless Sensor Networks

JIANG Wenxian, MIAO Haixing, WANG Tian, WANG Cheng, LAI Yongxuan, LIANG Junbin

2017, 30(6): 1216-1223. doi: 10.3969/j.issn.0258-2724.2017.06.024

[Abstract](473)

Abstract:
In order to prolong the lifetime of wireless sensor networks by using energy-rechargeable mobile devices, this paper designs a mobile energy replenishment method wherein a mobile element gathers data and recharges sensors simultaneously. Firstly, the whole sensor network is divided into several sub-regions equally and the sensors in each sub-region are formed into a cluster. Secondly, considering the energy in a whole cluster, the mobility path is designed to find the energy-efficient mobile trace of the mobile element. Finally, in the simulation experiment, we used ten different random network topologies to show the comparisons with extensive simulation experiments under different velocities and deadlines. The results indicate that the proposed algorithm increases lifetime by approximately 6 000 s compared with Nearest-Job-Next with Pre-emption (NJNP) under the deadline of 800 s. Moreover, the proposed algorithm increases lifetime by approximately 14 000 s compared with NJNP at velocity of 5 m/s. Thus, the proposed algorithm can improve recharging efficiency and prolong the lifetime of wireless sensor networks, which can be used in large-scale sensor networks.

Heuristic Complete Algorithm for SAT Problem by Using Logical Deduction

CHEN Qingshan, XU Yang, HE Xingxing

2017, 30(6): 1224-1232. doi: 10.3969/j.issn.0258-2724.2017.06.025

[Abstract](444)

Abstract:
In order to address the issue of branch decision-making inefficiency in solving the satisfiability problem (SAT), a heuristic complete algorithm based on logical deduction group (LDG) is proposed. Specifically, the proposed algorithm chooses the remaining unsolved clauses to make the logical deduction such that an assignment sequence of partial satisfiability is obtained, further guiding the solver to first search its solution space. For the satisfiable problem, the partial assignment can be extended group by group to a complete assignment by iteratively utilizing the deductive process. For the unsatisfiable problem, it can be straightforward to judge whether there exists an empty clause. The international competition instance in SAT is adopted for comparison with the typical exponential variable state independent decaying sum (EVSIDS) decision heuristic. The experimental results demonstrate that LDG can solve 42 more problems than EVSIDS, and achieve 22.8% and 17.8% reductions in average solution time for the satisfiable and unsatisfiable problems, respectively, as well as a 20.1% reduction in total average time.

Application of Hidden Markov Model for Fault Analysis of ODO in Train Location Unit

ZANG Yu, SHANGGUAN Wei, ZHANG Junzheng, CAI Baigen, WANG Huashen

2017, 30(6): 1233-1240. doi: 10.3969/j.issn.0258-2724.2017.06.026

[Abstract](541)

Abstract:
During the running of a train, the lock-and-slip state of a wheel may occur and hence, an Odometers (ODO) will temporarily display the state of functional failure. To address this issue, fault analysis of ODO based on the application of hidden Markov model (HMM) has been carried out. Firstly, an algorithm based on neural networks (NNs) was applied for the fault diagnosis of ODO. A fault diagnosis method based on HMM was then studied from the perspective of pattern recognition. The observation of the quantization coding sequence was introduced into the HMM classifier for establishing the hidden fault states of the ODO model with the aid of training data. By entering the observation sequence matching with the HMM classifier, the effective state of ODO was obtained. Finally, genetic algorithms (GAs) were employed for obtaining the parameters of HMM, rather than employing the Baum-Welch (B-W) algorithm. The obtained case study results show that the fault identification accuracy based on NN is 33.3%, the diagnos-tic accuracy based on HMM can reach up to 100% between normal and fault conditions, and the overall diagnostic accuracy is 95%. The training speed of GA can help in reaching the steady state quickly and the training accuracy can be improved by 86%. Furthermore, the obtained results show that the classification ability of NN is comparatively weak and cannot meet the required needs of classification when there is high noise in the data. The GA can improve on the drawbacks of the B-W algorithm that in turn can easily fall into the local optimum. The fault diagnosis method based on HMM has higher accuracy when compared with the NN-based method.

Search Advance Search

Journal Search

Search Type

NewsMore

instruction for authors

Read this journal

LinksMore

2017 Vol. 30, No. 6