CN
EN
Adaptability of Composite Sleeper for Heavy Railway Steel Bridge
Oct 06, 2021
Abstract: Wooden sleepers are traditionally used to transmit live loads on railway steel bridges in China. Although wood has good elasticity and insulation properties, it has the disadvantages of large amount of wood and cumbersome maintenance and repair. Composite sleepers can be similar in shape and weight to wooden sleepers. Composite sleepers which can be designed in terms of shape and weight are similar as wooden sleepers. In order to study the applicability of the composite material sleeper on the 30 t axle mass load railway,coupling vibration analysis of the 64m single-wire railway underpinning with wooden sleeper and composite material sleeper was carried out. At different speeds through the bridge to examine the dynamic performance of train and bridge. The results show that:in the case of different types of sleeper, the vertical displacement and acceleration of the train are lower than the standard limit for the bridge, and the traffic safety requirements are satisfied. For the train, wheel load reduction and body vibration are also required to meet the requirements, and pavement or composite sleeper when the system dynamic response is no significant difference. The stress of the composite sleeper is larger than that of the wooden sleeper, but still far less than the allowable stress of the composite sleeper. The deformation of the composite sleeper is less than that of the sleeper, the deformation is less than permissible deformation;for the composite sleeper, although the load near the force point of the sleeper is bigger than the sleeper, but the gap between the two is less than 2%, basically the same. Therefore, composite sleepers can meet the train in the range of 60~90 km/h safe and stable operation requirements and suitable as a substitute for wooden sleepers.
Key words: Composite material sleeper, safety, stability, steel truss girder, train-track-bridge coupling vibration
Chinese railway steel bridges traditionally use wooden sleepers to transmit live loads, and the exposed deck is composed of guard rails, guard rails and related joints. Although wood has good elasticity and insulation properties, there are obvious shortcomings in long-term operation. Firstly, it needs large amount of wood, which requires about 0.4~0.5 m3 of wood per linear meter of bridge deck, and because of the short life cycle of wood, it needs to be replaced frequently, and the second is the connection. There are many parts, the connection method and the operation items are relatively cumbersome, and the mechanical wear or cracking leads to a short maintenance cycle, which greatly increases the maintenance and repair workload. Therefore, the new bridges have basically adopted concrete sleepers. However, for existing steel bridges, the significant difference in weight between concrete sleepers and wooden sleepers prevents concrete sleepers from being used as substitutes for wooden sleepers, while composite sleepers can be designed to be similar in shape and weight to wooden sleepers through cross-sectional design as a viable option.
Scholars in domestic and abroad have carried out a series of studies on composite sleepers. Sekisui Chemical Industry Co., Ltd. of Japan has developed a composite sleeper, focusing on its durability: Japan Railway Comprehensive Technology Research Institute tested the composite sleeper and verified its good fatigue performance: Graz University of Technology, Austria, Mingqiao The composite sleeper on the surface steel bridge has undergone a life-cycle cost (LCC) analysis: Kunihei University of Technology, Germany; the composite sleeper has been tested and studied according to European standards: the American Railway Transportation Technology Center monitors the performance of the composite sleeper, including sleeper deflection and the gauge, the University of Illinois in the United States studied high-density polyethylene sleepers according to the test methods recommended by the Railway Engineering and Maintenance Association. The results showed that under load they showed a brittle fracture failure mode: The Shuihua Bridge of Chengdu Railway in China uses synthetic sleepers instead of degraded wooden sleepers for track and observe its performance; Yu Shouchen from Beijing University of Chemical Technology uses waste media polymers to trial composite sleepers through a two-stage extruder. Chen Yuxiao, Northeast Forestry University, from the perspective of satisfying compressive strength Starting off, a new type of composite sleeper was designed. The previous research of domestic and foreign scholars paid more attention to the design of composite sleepers from the perspectives of materials, structure and technology, as well as the study of failure modes of composite sleepers, etc. There is a lack of research on the dynamic response of composite sleepers, especially on heavy railways. Adaptability research.
This paper takes the glass fiber reinforced resin composite sleepers (composite sleepers for short) laid on the exposed deck of the existing steel truss bridge as the research object, and compares and analyzes the original wooden sleepers. The adaptability of the above mainly includes two parts. The first part examines the impact of composite sleepers on the dynamic response of trains, lines and bridges after replacing wooden sleepers with composite sleepers through train-line-bridge dynamic simulation analysis. The second part compares and analyzes the stress and deformation of composite sleepers and wooden sleepers.
1 Train line-bridge dynamic simulation model
The material of the 64 m single-track steel truss girder bridge is 16Mnq, and the live load is Grade 22. The line on the original bridge uses wooden sleepers on the exposed bridge surface. Affected by the anti-climbing angle steel, the wooden sleepers on the bridge are not evenly arranged. Instead, 23 wooden sleepers are arranged every 8m between internodes, and two at each end of the bridge, totaling 188. Fourteen wooden sleepers were installed at the transition sections on both sides of the steel bridge, followed by type III concrete sleepers. The transition section is symmetrical. The arrangement of wooden sleepers at the ends and 8m internodes is shown in Figure 1.
.png)
In order to study the impact of the replacement of wooden sleepers with composite sleepers on the dynamic performance of the track bridge system, a dynamic analysis model of the track bridge was established for the 64m single-track bridge welded under truss girder bridge and the line, in which the line structure considered non-uniform layout In addition to the influence of sleeper changes, the length of the lines on both sides of the bridge must not be less than the sum of the train length and 20 times the sleeper spacing to eliminate the influence of the boundary effect of the rail ends. In actual calculations, the lines on both sides of the bridge are both 187.23m long. The train uses a 30 t axle mass heavy train.
The train consists of multiple locomotives and wagons. Each train is a multi-degree-of-freedom vertical vibration system composed of wagon body, bogie, wheelset, springs and damping. From the perspective of solving engineering problems, the following assumptions are made for the wagon: ①The train body, bogie and wheelset are all rigid bodies; ②The wagon moves at a constant velocity on the bridge, regardless of the influence of the longitudinal axis direction power; ③The damping between all suspension systems of trains is viscous damping, and all springs are linear springs: ④The normal force between the wheel and rail is determined by the Hertzian nonlinear elastic contact theory, allowing the wheel and rail to separate from each other.
Based on the characteristics of the suspension parameters of a 30 t axle heavy train, this paper adopts a two-series suspension train model with 10 degrees of freedom in total. The schematic diagram of the train model is shown in Figure 2. The calculation conditions are listed in Table 1.
Table 1 Summary of working conditions for coupled vibration analysis of train-line-bridge
The track structure is an important part of the train-line-bridge dynamics system. On the one hand, it bears the force exerted by the train and at the same time transmits the received force to the bridge, causing the bridge to vibrate and deform.
On the other hand, it transmits the vibration and deformation of the bridge to the train, which affects the dynamic performance of the train. In the dynamic analysis, the continuous distributed parameter multi-layer discrete point support beam model is used for the track structure, that is, the rail is regarded as an infinite Euler beam supported by continuous elastic discrete points, and the foundation under the rail is discrete along the longitudinal direction. The discretization takes each sleeper fulcrum as the basic element, and the rail and sleeper of each element and the sleeper and the bridge deck are connected by springs and dampers. Figure 3 shows the track dynamics model of the exposed bridge deck.
.png)
Figure 2 train model with secondary suspension
.png)
Figure 3 Track Dynamics Model of Open Bridge Deck
.png)
Figure 4 Cross section size of composite sleeper
The steel bridge is exposed and the wooden sleepers are connected by separate K-shaped fasteners. The composite sleepers are connected by elastic clip II fasteners. The line flexibility is mainly provided by the sleepers. The cross-sectional dimensions of the composite sleepers are shown in Figure 4. In the dynamic performance analysis, the rail and sleeper are elastically connected. The sleeper is simulated as a rigid body with mass, and its elastic parts are expressed by the elastic connection between the sleeper and the bridge deck. Existing resources cannot directly provide the elasticity of wooden sleepers and composite sleepers. For this reason, the finite element software MIDAS modeling analysis is used to determine their equivalent stiffness. Both composite sleepers and wooden sleepers are modeled by solid elements. The sleepers and steel beams are connected in the form of simple support within the range of their contact surfaces. The surface load is applied within the corresponding range of the iron pad. The equivalent stiffness of the sleepers and composite sleepers is calculated. The schematic diagram is shown in Figure 5. The equivalent stiffness of the wooden sleepers on the bridge is 170MN/m, and the equivalent stiffness of the composite sleepers is 187MN/m.
.png)
.png)
See Table 2 for bridge sleeper materials and section characteristics.
Table 2 Bridge sleeper materials and section characteristics
Track irregularity is the main source of vibration in the wheel-rail system. It is the control factor that affects the safety and smoothness of driving. The influence of track irregularity on the dynamic characteristics of the wheel-rail system is reflected by the change of the wheel-rail contact relationship. Since the American pentad spectrum is roughly equivalent to the track spectrum of my country’s three major trunk lines, the speed of trains can reach 128 km/h and the speed of passenger cars can reach 144 km/h. Therefore, the time-domain samples of track irregularities generated by the American pentad spectrum are used in the calculation. The maximum amplitude of irregularities samples are listed in Table 3.
Table 3 Amplitude of track irregularity sample
mm
The change of the track sample with distance is shown in Figures 6 and 7.
.png)
Distance/km
Figure 6 U.S. five-level spectrum left track irregularities
.png)
Distance/km
Figure 7 U.S. five-level spectrum left track irregularities
Each member of the bridge is modeled by spatial beam element. For the second-stage load of the bridge, it is distributed as a uniform load mass to the railway longitudinal beams and lower chords. The bridge structure adopts uniform stiffness matrix and uniform mass moment order, and the damping matrix adopts the Rayleigh damping model, which is expressed as a combination of mass matrix and stiffness matrix by the Rayleigh damping coefficient. The mathematical expression is that the steel bridge damping ratio is 1% Select. The finite element model of the bridge is shown in Figure 8. In order to more accurately simulate the influence of the bridge deck structure on the power of the train bridge, the connecting rod between the two longitudinal beams is considered, as shown in Figure 9.
.png)
Figure 8 Finite element model of bridge
.png)
Figure 9 Finite element model of bridge deck
The train-line-bridge system is bounded by steel rails, above which is the train subsystem, and below it is the rail and bridge subsystem. The coupling relationship between the train system and the rail-bridge system is reflected in the wheel-rail force between the wheel and the rail. This article uses Hertzian nonlinear elastic contact theory to describe wheel-rail contact springs. According to the Hertz theory, the relationship between the wheel-rail vertical force Nz(t) and the elastic compression between the wheel and rail is as follows
.png)
Where G--wheel-rail contact constant (m/N2/3).
δZ(t)--Elastic compression between wheel and rail (m)
oZ(rr-Elastic compression between the wheel rails (m)
After establishing the motion differential equation of the lane-bridge system, the Newmark-β method is used to solve it numerically.
2 Train-line-bridge vertical coupling dynamics analysis
2.1 Bridge natural frequency
The calculation results of the first three natural frequencies of the steel bridge calculated by the MIDAS software are shown in Table 4, and the vibration modes are shown in Figures 10~12.
Table 4 The first five natural frequencies of steel bridges
Hz
The first three modes of the bridge are as follows:
.png)
Figure 10 The first mode shape
.png)
Figure 11 The second mode shape
.png)
Figure 12 The third mode shape
2.2 System dynamic response analysis
In order to fully analyze whether the composite sleepers meet the safety requirements, the dynamic response of the train-line bridge under the condition of using wooden sleepers and composite sleepers is calculated. The calculation results are shown in Tables 5~8. When composite sleepers are installed, the response time history curves of trains, lines and bridges are shown in Figures 13-18. According to China's "Railway train Dynamics Performance Evaluation and Test Appraisal Specification" (GB 5599-I 985), the allowable limit of wheel load reduction rate in the safety index is not more than 0.6; the vertical vibration acceleration of the freight train body in the running smoothness index The limit value is a≤0.7g; referring to the "Code for Verification of Railway Bridges and Culverts" and UIC specifications, for ballastless tracks, the vertical acceleration of the bridge deck amax≤0.5g.
Key words: Composite material sleeper, safety, stability, steel truss girder, train-track-bridge coupling vibration
Chinese railway steel bridges traditionally use wooden sleepers to transmit live loads, and the exposed deck is composed of guard rails, guard rails and related joints. Although wood has good elasticity and insulation properties, there are obvious shortcomings in long-term operation. Firstly, it needs large amount of wood, which requires about 0.4~0.5 m3 of wood per linear meter of bridge deck, and because of the short life cycle of wood, it needs to be replaced frequently, and the second is the connection. There are many parts, the connection method and the operation items are relatively cumbersome, and the mechanical wear or cracking leads to a short maintenance cycle, which greatly increases the maintenance and repair workload. Therefore, the new bridges have basically adopted concrete sleepers. However, for existing steel bridges, the significant difference in weight between concrete sleepers and wooden sleepers prevents concrete sleepers from being used as substitutes for wooden sleepers, while composite sleepers can be designed to be similar in shape and weight to wooden sleepers through cross-sectional design as a viable option.
Scholars in domestic and abroad have carried out a series of studies on composite sleepers. Sekisui Chemical Industry Co., Ltd. of Japan has developed a composite sleeper, focusing on its durability: Japan Railway Comprehensive Technology Research Institute tested the composite sleeper and verified its good fatigue performance: Graz University of Technology, Austria, Mingqiao The composite sleeper on the surface steel bridge has undergone a life-cycle cost (LCC) analysis: Kunihei University of Technology, Germany; the composite sleeper has been tested and studied according to European standards: the American Railway Transportation Technology Center monitors the performance of the composite sleeper, including sleeper deflection and the gauge, the University of Illinois in the United States studied high-density polyethylene sleepers according to the test methods recommended by the Railway Engineering and Maintenance Association. The results showed that under load they showed a brittle fracture failure mode: The Shuihua Bridge of Chengdu Railway in China uses synthetic sleepers instead of degraded wooden sleepers for track and observe its performance; Yu Shouchen from Beijing University of Chemical Technology uses waste media polymers to trial composite sleepers through a two-stage extruder. Chen Yuxiao, Northeast Forestry University, from the perspective of satisfying compressive strength Starting off, a new type of composite sleeper was designed. The previous research of domestic and foreign scholars paid more attention to the design of composite sleepers from the perspectives of materials, structure and technology, as well as the study of failure modes of composite sleepers, etc. There is a lack of research on the dynamic response of composite sleepers, especially on heavy railways. Adaptability research.
This paper takes the glass fiber reinforced resin composite sleepers (composite sleepers for short) laid on the exposed deck of the existing steel truss bridge as the research object, and compares and analyzes the original wooden sleepers. The adaptability of the above mainly includes two parts. The first part examines the impact of composite sleepers on the dynamic response of trains, lines and bridges after replacing wooden sleepers with composite sleepers through train-line-bridge dynamic simulation analysis. The second part compares and analyzes the stress and deformation of composite sleepers and wooden sleepers.
1 Train line-bridge dynamic simulation model
The material of the 64 m single-track steel truss girder bridge is 16Mnq, and the live load is Grade 22. The line on the original bridge uses wooden sleepers on the exposed bridge surface. Affected by the anti-climbing angle steel, the wooden sleepers on the bridge are not evenly arranged. Instead, 23 wooden sleepers are arranged every 8m between internodes, and two at each end of the bridge, totaling 188. Fourteen wooden sleepers were installed at the transition sections on both sides of the steel bridge, followed by type III concrete sleepers. The transition section is symmetrical. The arrangement of wooden sleepers at the ends and 8m internodes is shown in Figure 1.
In order to study the impact of the replacement of wooden sleepers with composite sleepers on the dynamic performance of the track bridge system, a dynamic analysis model of the track bridge was established for the 64m single-track bridge welded under truss girder bridge and the line, in which the line structure considered non-uniform layout In addition to the influence of sleeper changes, the length of the lines on both sides of the bridge must not be less than the sum of the train length and 20 times the sleeper spacing to eliminate the influence of the boundary effect of the rail ends. In actual calculations, the lines on both sides of the bridge are both 187.23m long. The train uses a 30 t axle mass heavy train.
The train consists of multiple locomotives and wagons. Each train is a multi-degree-of-freedom vertical vibration system composed of wagon body, bogie, wheelset, springs and damping. From the perspective of solving engineering problems, the following assumptions are made for the wagon: ①The train body, bogie and wheelset are all rigid bodies; ②The wagon moves at a constant velocity on the bridge, regardless of the influence of the longitudinal axis direction power; ③The damping between all suspension systems of trains is viscous damping, and all springs are linear springs: ④The normal force between the wheel and rail is determined by the Hertzian nonlinear elastic contact theory, allowing the wheel and rail to separate from each other.
Based on the characteristics of the suspension parameters of a 30 t axle heavy train, this paper adopts a two-series suspension train model with 10 degrees of freedom in total. The schematic diagram of the train model is shown in Figure 2. The calculation conditions are listed in Table 1.
Table 1 Summary of working conditions for coupled vibration analysis of train-line-bridge
| Item | Train throughput | Sleeper | Calculated train speed/（km/h） | Irregular tracks |
| Heavy train with 30t with axle mass | 10 train throughput | Wooden sleeper, Composite sleeper |
60,70, 80,90 |
American pentagram sample |
On the other hand, it transmits the vibration and deformation of the bridge to the train, which affects the dynamic performance of the train. In the dynamic analysis, the continuous distributed parameter multi-layer discrete point support beam model is used for the track structure, that is, the rail is regarded as an infinite Euler beam supported by continuous elastic discrete points, and the foundation under the rail is discrete along the longitudinal direction. The discretization takes each sleeper fulcrum as the basic element, and the rail and sleeper of each element and the sleeper and the bridge deck are connected by springs and dampers. Figure 3 shows the track dynamics model of the exposed bridge deck.
Figure 2 train model with secondary suspension
Figure 3 Track Dynamics Model of Open Bridge Deck
Figure 4 Cross section size of composite sleeper
The steel bridge is exposed and the wooden sleepers are connected by separate K-shaped fasteners. The composite sleepers are connected by elastic clip II fasteners. The line flexibility is mainly provided by the sleepers. The cross-sectional dimensions of the composite sleepers are shown in Figure 4. In the dynamic performance analysis, the rail and sleeper are elastically connected. The sleeper is simulated as a rigid body with mass, and its elastic parts are expressed by the elastic connection between the sleeper and the bridge deck. Existing resources cannot directly provide the elasticity of wooden sleepers and composite sleepers. For this reason, the finite element software MIDAS modeling analysis is used to determine their equivalent stiffness. Both composite sleepers and wooden sleepers are modeled by solid elements. The sleepers and steel beams are connected in the form of simple support within the range of their contact surfaces. The surface load is applied within the corresponding range of the iron pad. The equivalent stiffness of the sleepers and composite sleepers is calculated. The schematic diagram is shown in Figure 5. The equivalent stiffness of the wooden sleepers on the bridge is 170MN/m, and the equivalent stiffness of the composite sleepers is 187MN/m.
- Wooden sleeper
- Composite sleeper
See Table 2 for bridge sleeper materials and section characteristics.
Table 2 Bridge sleeper materials and section characteristics
| Bridge sleeper type | Wooden sleeper | Composite sleeper | |
| Outside size(mm x mm x mm) | 3000x220x260 | 3000x240x170 | |
| Bending elastic modulus E/GP | 9.0 | 40.0 | |
| Equivalent stiffness/(N/m) | 1.7x108 | 1.87x108 | |
| Moment of inertia of section/m4 | Torque J | 4.519x10-4 | 5.966x10-5 |
| Around the horizontal axis I1 | 3.222x10-4 | 5.519x10-4 | |
| Around the vertical axis I2 | 2.307x10-4 | 5.947x10-4 | |
Track irregularity is the main source of vibration in the wheel-rail system. It is the control factor that affects the safety and smoothness of driving. The influence of track irregularity on the dynamic characteristics of the wheel-rail system is reflected by the change of the wheel-rail contact relationship. Since the American pentad spectrum is roughly equivalent to the track spectrum of my country’s three major trunk lines, the speed of trains can reach 128 km/h and the speed of passenger cars can reach 144 km/h. Therefore, the time-domain samples of track irregularities generated by the American pentad spectrum are used in the calculation. The maximum amplitude of irregularities samples are listed in Table 3.
Table 3 Amplitude of track irregularity sample
mm
| Irregularity of the left rail | Irregularity of the right rail |
The left rail is uneven |
The right rail is uneven |
|
| U.S. five-level spectrum | 10.357 | 10.6873 | 13.7322 | 14.8668 |
Distance/km
Figure 6 U.S. five-level spectrum left track irregularities
Distance/km
Figure 7 U.S. five-level spectrum left track irregularities
Each member of the bridge is modeled by spatial beam element. For the second-stage load of the bridge, it is distributed as a uniform load mass to the railway longitudinal beams and lower chords. The bridge structure adopts uniform stiffness matrix and uniform mass moment order, and the damping matrix adopts the Rayleigh damping model, which is expressed as a combination of mass matrix and stiffness matrix by the Rayleigh damping coefficient. The mathematical expression is that the steel bridge damping ratio is 1% Select. The finite element model of the bridge is shown in Figure 8. In order to more accurately simulate the influence of the bridge deck structure on the power of the train bridge, the connecting rod between the two longitudinal beams is considered, as shown in Figure 9.
Figure 8 Finite element model of bridge
Figure 9 Finite element model of bridge deck
The train-line-bridge system is bounded by steel rails, above which is the train subsystem, and below it is the rail and bridge subsystem. The coupling relationship between the train system and the rail-bridge system is reflected in the wheel-rail force between the wheel and the rail. This article uses Hertzian nonlinear elastic contact theory to describe wheel-rail contact springs. According to the Hertz theory, the relationship between the wheel-rail vertical force Nz(t) and the elastic compression between the wheel and rail is as follows
Where G--wheel-rail contact constant (m/N2/3).
δZ(t)--Elastic compression between wheel and rail (m)
oZ(rr-Elastic compression between the wheel rails (m)
After establishing the motion differential equation of the lane-bridge system, the Newmark-β method is used to solve it numerically.
2 Train-line-bridge vertical coupling dynamics analysis
2.1 Bridge natural frequency
The calculation results of the first three natural frequencies of the steel bridge calculated by the MIDAS software are shown in Table 4, and the vibration modes are shown in Figures 10~12.
Table 4 The first five natural frequencies of steel bridges
Hz
| First level | Second level | Third level | |
| Natural frequency | 1.809 | 3.179 | 3.928 |
| Mode shape | Symmetrical bending | Cross bend + twist | Symmetrical vertical bend |
The first three modes of the bridge are as follows:
Figure 10 The first mode shape
Figure 11 The second mode shape
Figure 12 The third mode shape
2.2 System dynamic response analysis
In order to fully analyze whether the composite sleepers meet the safety requirements, the dynamic response of the train-line bridge under the condition of using wooden sleepers and composite sleepers is calculated. The calculation results are shown in Tables 5~8. When composite sleepers are installed, the response time history curves of trains, lines and bridges are shown in Figures 13-18. According to China's "Railway train Dynamics Performance Evaluation and Test Appraisal Specification" (GB 5599-I 985), the allowable limit of wheel load reduction rate in the safety index is not more than 0.6; the vertical vibration acceleration of the freight train body in the running smoothness index The limit value is a≤0.7g; referring to the "Code for Verification of Railway Bridges and Culverts" and UIC specifications, for ballastless tracks, the vertical acceleration of the bridge deck amax≤0.5g.