# An Investigation of a Locomotive Structural Crashworthiness Using Finite Element Simulation.

IntroductionIndia is a developing country with the second largest population of 1.27 billion people. Indian economy is expected to grow at 5-8% gross domestic product (GDP) in the next decade. This growth in the economy will result in further expansion of metros and Tier 2 cities [1]. Indian Railways (IR) is the fourth largest railway network in terms of track, passengers, and cargo volume handling. IR will have to cater the increasing demands of passengers for faster and safe travel. High-speed trains are required to meet this purpose. Therefore, IR is focusing on developing new high-speed tracks and upgrading the existing infrastructure to exploit its higher speed limits, that is, from the existing 130-160 km/hr to 200 km/hr on major routes [2, 3, 4, 5, 6]. Hence to achieve this, IR has to attain service quality of international standards, considering safety parameters as indispensable. Crashworthiness, which is the structural ability of a vehicle to protect the passengers and drivers during an impact, is considered to be the important parameter at high speed [7, 8, 9, 10].

Crashworthiness is evaluated either predictively using finite element models and experiments or reflectively by breaking down crash results. Different criteria are utilized to evaluate crashworthiness, including the acceleration response and deformation patterns of the vehicle structure [11, 12, 13]. For the analysis, four scenarios are taken into account out of which two are associated with the collision of rail equipment with each other and the other two are related to the collision of the train to road vehicles. Practical tests using full-scale model and computer simulations are performed to study the effects of the collision on rail vehicles [14, 15, 16, 17]. The collision performance of rail vehicle is usually conducted with symmetric rigid objects. The Technical Specification for Interoperability (TSI) is a standard procedure where collisions with rigid objects are conducted to evaluate the crashworthiness of rail vehicle [19]. Computer simulation is the preferred method to test the crashworthiness as compared to the practical method of testing which not only is an expensive procedure but also consumes a lot of time.

Finite element analysis (FEA) and multibody dynamics methods are currently utilized to examine rail vehicle collisions, crashworthiness, and deformations of energy-absorbing elements. Multibody dynamics is used to study the complete movement of a whole train, whereas finite element modeling is performed to obtain refined outcomes [18, 19, 20, 21]. The FEA is a numerical method which breaks down the continuous system to discrete systems which are called elements. FEA is a powerful method of solving challenging engineering problems because it permits applying boundary conditions to make an accurate model which allows the study of collision impacts. Explicit FEA is used to simulate collision of rail vehicles as the time frame of such events is very small. It involves nonlinear behavior and deformations that occur commonly in a collision scenario.

IR presently uses electric and diesel locomotives. For electrical passenger locomotive, Wide gauge Alternating current for Passenger (WAP) series, that is, WAP-1, WAP-4, WAP-5, and WAP-7, etc., is used which run through 25 kV AC system with overhead lines. In this article, the analysis is mainly focused on the modification of the WAP-5 locomotive design. Therefore, it is essential to examine the locomotive crash behavior during the impact test. In this FEA, the crashworthiness of a locomotive is assessed by impacting it on a rigid wall at various speeds, that is, 25 km/hr, 100 km/hr, 160 km/hr, and 225 km/hr.

Finite Element Model for Locomotive

Description of Locomotive

WAP-5 is a dual-cab forward-type high-speed IR electric locomotive. It was initially imported from ASEA Brown Boveri, Switzerland, but is now manufactured by Chittaranjan Locomotive Works, India. It has a peak power output of 4500 kW and includes regenerative braking as one of its eminent characteristics. The other main features are flexible gear coupling, wheel-mounted disc brakes, pantry loads, and an ability to attain higher speed up to 225 km/hr.

Braking system includes 160 kN regenerative brakes, automatic train air brakes, locomotive disc brakes, and a parking brake. It is the fastest locomotive of IR and runs at 160 km/hr.

Figures 1 and 2 show the geometrical diagram and a side view of WAP-5 locomotive, respectively. Salient features of WAP-5 are listed in Table 1.

Finite Element Model

The FE techniques have been thoroughly utilized in simulating diverse dynamic problems through the discretization of the whole structure into finite elements. Implicit and explicit methods are used to solve the FEA. For crash simulation events which have a nonlinear transient dynamic problem and have very short duration of time, an explicit method is the most suitable one [8]. The locomotive and its rigid wall have been modeled using solid works; meshing is done in hyper works and imported to ANSYS for explicit analysis (Figure 3).

The FE model is based on an actual representation of the locomotive front structure which had a fine mesh and treated as a deformable component. Different sections of the locomotive are simplified and considered as rigid bodies except the extreme modules (Appendix A). Skin plates are neglected in the modeling. The complete locomotive is generated including substructures such as pushback couplers and buffers.

The coupler with draft gear works according to the force-stroke characteristics [12]. Buffers absorb a large amount of the impact energy during the crash. However, the body front structure is also subjected to the impact as the buffers are mounted to the frame. Computer-aided design (CAD) model is based on solid elements. The actual front structure is generally made of a large number of steel elements in the form of sheet metal plates and sections welded together; therefore, it is modeled with shell finite elements. The transformation from solid objects (CAD model) to the surface model (FEM) required applying of the midsurface procedure (Figure 4). Midsurface is generated between two sidewalk of the solid exactly at the mid of them.

The configuration of the frame is very complex (fillets, sections, etc.) and automatic meshing is utilized. The FE mesh of the frontal locomotive is gradually refined toward the impacting end of the model to exactly determine the deformation of the structure in a collision, with a characteristic element length of 15 mm. The mesh is relatively regular and it consists of both rectangular and triangular elements. The anti-climbers have a characteristic element length of 0.5 mm. These models are formulated to determine their crushing and vertical strength. Large deformation is anticipated in the locomotive frontal area and it has an element size of approximately 5 mm. The element size increases to 25 mm for the remaining part and increases progressively toward the rear of the locomotive.

The front structure is considered as a deformable component and *MAT_PLASTIC_KINEMATIC is applied to the whole frame elements. This model is suitable for incorporating isotropic and kinematic hardening plasticity. Table 2 gives the material properties of WAP-5. Complete FE model (Figure 5) consists of 150,367 finite elements and 249,285 nodes. 98,950 deformable shell elements are used to simulate the frame and buffers, whereas 21,586 solid elements are used to simulate other components as rigid bodies. 28 beam elements are used to represent springs and dampers. In addition, 40-lumped masses are attached to the model to ensure its correct mass. Meshing details are provided in Table 3.

The locomotive frame consists of over 94 thousands of grouped elements in nine parts, and each part is described by the different thickness of the shell element (Appendix A). As the sum of different masses considered in the model represents the total mass of the designed locomotive (about 11 tons), the density of some components is modified and few lumped parameters are added in the model to ensure its proper mass.

A mesh quality is checked through orthogonal and skewness quality and mesh metric is given in Table 4. The element percentage lie in different quality zone is given in Table 5. It can be seen that the major percentage of elements lie in very good zone and very few percentage of elements fall in unacceptable zone. Therefore, the mesh quality has passed the orthogonal and skewness quality test.

Validation of Crashworthiness

The FE model is validated in terms of acceleration and energy balance by experimental result obtained by the Research Designs and Standards Organisation (RDSO) and European Standard (EN) 15227 [22]. The head-on collision analysis with the identical unit is considered and locomotive is impacted on a rigid wall at the speed of 25 km/hr. Since the rigid wall does not absorb any energy, it results in a condition of a head-on collision of two trains and satisfies the first condition as per EN 15227.

The simulation results show that the frontal part of locomotive crashed and maximum deformation is about 2.01 m as shown in Figure 6.

Figure 7 shows the comparison of acceleration during collision. To determine the efficiency of the crash analysis, it is essential to evaluate the energy balance of the system. The energy balance estimate in the crash analysis is expressed in Equation 1:

[E.sub.I] + [E.sub.FD] + [E.sub.KE] - [E.sub.SW] + [E.sub.v] = [E.sub.TOT] Eq. (1)

Figure 8 indicates the total energy segment [E.sub.TOT], which have been determined from Equation 1, and few major energy components, that is, kinetic energy [E.sub.KE], internal energy [E.sub.I], and artificial strain energy, external work by mass scaling, and work done by the external loads during the collision analysis. If the total energy in the finite element model does not change, the total error is usually below 1%, which indicates that the crash analysis results have been validated. It is obvious that kinetic energy converts into internal strain energy in the course of simulation. As per ECE R66 norms, the amount of nonphysical energy components must be within 5% of the total energy. The artificial strain energy, which consists of the stored energy in the hourglass resistances and transverse shear in the shell elements, is negligible as compared to the total energy which is in the acceptable range. The values of acceleration are in the ranges as per experimental result. There is a good matching among the experimental and simulated results.

Results and Discussion

Simulation Scenario

The simulation scenario of a locomotive impacting on a rigid wall (Figure 5) is a general representative scenario of crash simulation. Three scenarios are taken into consideration with the initial speed of 100 km/hr (average speed), 160 km/hr (maximum running speed), and 225 km/hr (design speed) [22, 231. The simulation is performed for 0.4 s, which is sufficient to record the collision response. A controlled gradual collapse of the frontal structure is crucial for locomotive crash-worthiness design to absorb energy [24, 25, 26, 27, 28, 29, 30]. To display the deformation clearly, the second half of the locomotive is neglected. Initially, the front side sill, absorber, and front underframe have impacted with the rigid wall and collapsed, absorbing most of the total energy (Figure 9); then, the cab begins to deform and absorbs the remaining energy.

For the speed of 100 km/hr and 160 km/hr, the maximum deformation is 4.07 m and 6.05 m, respectively, and the front cabin of the locomotive is crashed completely. The deformation sequence of the front-end structure of a locomotive during the crash (160 km/hr) is described in Figure 10. The force history curves for the collision are described in Figure 11.

The first event depicts energy absorbed by the front coupler, after which the second event depicts forces occurring at collision energy-absorbed devices. Impact force is having high values of thrust on the underframe and the upper frames, which progressively increased. Collision energy absorption scenarios for frontal crashworthiness of the locomotive are developed corresponding to the design intent and at the speed of 225 km/hr; the maximum deformation is 8.22 m (Figure 12).

During this impact, the locomotive is crashed approximately to half of its length. The front cabin and middle section of locomotive are deformed completely. It is noticed that local buckling occurs at different points, which reduces the desired progressive damage behavior.

Reference node displacement (RND) in the longitudinal direction is given in Table 6. Figure 13 shows the RND in the longitudinal direction. The difference in the curve is for RND of the central area to the right and left sides of the locomotive, expressing the deformation of the central area to the right side of the locomotive and central area to the left side of the locomotive, respectively. It exhibits small rigid movement in the last stage, in which the right side moves forward and the left side retreats.

Conclusion

In this article, the main focus is on modernization of the WAP-5 locomotive. Therefore, it is required to determine the locomotive crash behavior during the impact test. The crashworthiness of a locomotive is carried out through the FE A by impacting it on a rigid wall to evaluate its maximum deformation at higher speed. FE model of the locomotive includes some simplifications due to high complexity of the actual object. The proposed FE model used in the dynamic numerical analysis is the only possibility since the experimental test on the complete locomotive is impractical and impossible at the moment. The FE model is validated at the speed of 25 km/hr in terms of acceleration and energy balance by experimental result and EN 15227. Longitudinal deceleration for the locomotive does not exceed the permissible values.

The buffers are fully compressed immediately after impacting with the wall and, therefore, are not able to absorb the energy. Thus, the energy must be absorbed by the frame that may lead to excessive strains and permanent plastic deformations. The results provide qualitative and quantitative data associated with the locomotive crash behavior during the collision. It may be useful for the designer in further modernization of the locomotive structure. This is the simplest and the most ideal crash scenario, but it is very useful to obtain general characteristics of the crash behavior of the locomotive. In the future, the crash study of two locomotives can be performed on the basis of the models presented.

References

[1.] Indian Cities, "Classification of Indian Cities," accessed May 2018, https://en.wikipedia.org/wiki/Classification_of_Indian_cities.

[2.] Sharma, S.K., Sharma, R.C., Kumar, A., and Palli, S., "Challenges in Rail Vehicle-Track Modeling and Simulation," International journal of Vehicle Structures and Systems 7(1):1-9, 2015, doi:10.4273/ijvss.71.01.

[3.] Sharma, S.K. and Kumar, A., "Dynamics Analysis of Wheel Rail Contact Using FEA," Procedia Engineering 144:1119-1128, 2016, doi:10.1016/j.proeng.2016.05.076.

[4.] Sharma, S.K. and Chaturvedi, S., "Jerk Analysis in Rail Vehicle Dynamics," Perspectives in Science 8, 2016, doi:648-650. doi:10.1016/i.pisc.2016.06.047

[5.] Sharma, R.C., Palli, S., Sharma, S.K., and Roy, M., "Modernization of Railway Track with Composite Sleepers," International Journal of Vehicle Structures and Systems 9(5):321-329, 2017, doi:10.4273/ijvss.9.5.10.

[6.] Sharma, S.K. and Kumar, A., "Ride Comfort of a Higher Speed Rail Vehicle Using a Magnetorheological Suspension System," Proceedings of the Institution of Mechanical Engineers, Part K: Journal of Multi-body Dynamics 232(1):32-48, 2018, doi:10.1177/1464419317706873.

[7.] Liana, P. and Richard, S., "Preliminary Development of Locomotive Crashworthy Components," Proceedings of the ASME/ASCE/IEEE 2011 Joint Rail Conference, 2011, 1-10, http://ntl.bts.gov/lib/37000/37900/37915/TR_C2011-56104_Locomotive_Crashworthiness.pdf.

[8.] Liana, P., Richard, S., and Simpson, G., "Finite Element Analysis and Full-Scale Testing of Locomotive Crashworthy Components," Proceedings of the ASME/ASCE/IEEE 2011 Joint Rail Conference, 2016, 1-11.

[9.] Lu, Z., Li, B., Yang, C., Zhao, H. et al., "Numerical and Experimental Study on the Design Strategy of a New Collapse Zone Structure for Railway Vehicles," International Journal of Crashworthiness 22(5):488-502, 2017, doi:10.1080/13588265.2017.1281080.

[10.] Naravane, A., Deb, A., Shivakumar, N.D. et al, "A Comparative Study of Composite and Steel Front Rails for Vehicle Front Impact Safety," International Mobility Congress and Exposition, The Automotive Research Association of India, Chennai, India, 2005, doi:10.4271/2005-26-324.

[11.] Sharma, S.K. and Kumar, A., "Disturbance Rejection and Force-Tracking Controller of Nonlinear Lateral Vibrations in Passenger Rail Vehicle Using Magnetorheological Fluid Damper," Journal of Intelligent Material Systems and Structures 29(2):279-297, 2018, doi:10.1177/1045389X17721051.

[12.] Sharma, S.K. and Kumar, A., "Impact of Electric Locomotive Traction of the Passenger Vehicle Ride Quality in Longitudinal Train Dynamics in the Context of Indian Railways," Mechanics and Industry 18(2), 2017, doi:10.1051/meca/2016047

[13.] Sharma, S.K. and Kumar, A., "Ride Performance of a High Speed Rail Vehicle Using Controlled Semi Active Suspension System," Smart Materials and Structures 26(5):055026, 2017, doi:10.1088/1361-665X/aa68f7.

[14.] Georgiou, G. and Zeguer, T., "On the Assessment of the Macro-Element Methodology for Full Vehicle Crashworthiness Analysis," International Journal of Crashworthiness 23(3):336-353, 2017, doi:10.1080/13588265.20171328723.

[15.] Kiani, M. and Yildiz, A.R., "A Comparative Study of Non-traditional Methods for Vehicle Crashworthiness and NVH Optimization," Archives of Computational Methods in Engineering 23(4):723-734, 2016, doi:10.1007/s11831-015-9155-y.

[16.] Lin, C., Gao, F., and Bai, Y., "Multiobjective Reliability-Based Design Optimisation for Front Structure of an Electric Vehicle Using Hybrid Metamodel Accuracy Improvement Strategy-Based Probabilistic Sufficiency Factor Method," International Journal of Crashworthiness 1-12, 2017, doi:10.1080/13588265.20171317466.

[17.] Kawasaki, T., Yamaguchi, T., and Mochida, T., "Railway-Vehicle Technologies for European Railways," Hitachi Review 57(1):61-65, 2008.

[18.] Zhou, H., Zhang, J., and Hecht, M., "Three-Dimensional Derailment Analysis of Crashed Freight Trains," Vehicle System Dynamics 52(3):37-41, 2014, doi:10.1080/00423114.2014.881512.

[19.] Zhang, Y, Xu, X., Lu, M., Hu, Z. et al., "Enhance Crashworthiness of Composite Structures Using Gradient Honeycomb Material," International Journal of Crashworthiness 1-12, 2017, doi:10.1080/13588265.20171367355.

[20.] Vani, A. and Jayachandraiah, B., "Crashworthiness of a Railway Vehicle to Reduce Overriding Effect by Using Abaqus Software," International Journal of Engineering Science Invention 4(7):14-22, 2015.

[21.] Spirka, S., Kemka, V., Kepka, M., and Malkovsk, Z., "Design of a Large Deformable Obstacle for Railway Crash Simulations According to the Applicable Standard," Applied and Computational Mechanics 6:83-92, 2012.

[22.] Bhattacharya, B.C., "Braking and Coupler Force Trails on 18 LHB A.C Coach Train Hauled by Single WAP5 A.C Electrical Locomotive of Northern Railways," Lucknow, India, 2001.

[23.] Indian Railways, "WAP-7 and WAP-5, Tractive Effort/Braking Effort in Technical Data and Notes," India, accessed May 2018, http://www.irfca.org/docs/index.html#tech.

[24.] Sharma, S.K., "Multibody Analysis of Longitudinal Train Dynamics on the Passenger Ride Performance due to Brake Application," Proceedings of the Institution of Mechanical Engineers, Part K: Journal of Multi-body Dynamic, First online, 2018, doi:10.1177/1464419318788775.

[25.] Sharma, S.K. and Kumar, A., "Impact of Longitudinal Train Dynamics on Ride Comfort and Train Operations: A Simulation-Based Study," Journal of Vibration Engineering and Technologies 6(3):197-203, 2018, doi:10.1007/s42417-018-0033-4.

[26.] Sharma, R.C. and Sharma, S.K., "Sensitivity Analysis of Three-Wheel Vehicle's Suspension Parameters Influencing Ride Behavior," Noise & Vibration Worldwide 49(7-8):272-280, 2018, doi:10.1177/0957456518796846.

[27.] Sharma, S.K. and Sharma, R.C, "Simulation of Quarter-Car Model with Magnetorheological Dampers for Ride Quality Improvement," International journal of Vehicle Structures and Systems 10(3):169-173, 2018, doi:10.4273/ijvss.l0.3.03.

[28.] Sharma, R.C. and Sharma, S.K., "Rail Vehicle Modelling and Simulation Using Lagrangian Method," International journal of Vehicle Structures and Systems 10(3):188-194, 2018, doi:10.4273/ijvss.l0.3.07.

[29.] Palli, S., Koona, R., Sharma, S.K., and Sharma, R.C., "A Review on Dynamic Analysis of Rail Vehicle Coach," International Journal of Vehicle Structures and Systems 10(3):204-211, 2018, doi:10.4273/ijvss.l0.3.10.

[30.] Sharma, S.K. and Kumar, A., "The Impact of Rigid-Flexible System on the Ride Quality of Passenger Bogies Using a Flexible Carbody," The Third International Conference on Railway Technology: Research, Development and Maintenance, Civil-Comp Press, Stirlingshire, UK, Paper 87, 2016, doi:10.4203/ccp.110.87.

Appendix A: Detailed Summary of the FE Model Part ID Part name Element type Number of elements 1 Buffers Shell 2506 2 Frame_floor Shell 1537 3 Frame 18 Shell 141,239 4 Frame 14 Shell 6425 5 Frame 16 Shell 16,853 6 Frame 11 Shell 17,521 7 Frame 12 Shell 8320 8 Frame 13 Shell 3648 9 Frame 15 Shell 11,716 10 Frame 17 Shell 198 11 Connectors Shell 530 12 Buffer spring Beam 5 13 Reservoir 11 Solid 140 14 Reservoir 22 Solid 374 15 Reservoir 33 Solid 250 16 Generator module Solid 83 17 Storage module Solid 98 18 Cab Solid 198 19 Electric module Solid 78 20 Pneumatic module Solid 52 21 Compressor module Solid 234 22 Other Solid 37,281 Total 249,285 Part ID Material type Thickness (shell only) (mm) 1 Plastic kinematic 25.0 2 Plastic kinematic 5.0 3 Plastic kinematic 5.0 4 Plastic kinematic 15.0 5 Plastic kinematic 10.0 6 Plastic kinematic 9.0 7 Plastic kinematic 7.6 8 Plastic kinematic 6.0 9 Plastic kinematic 20.0 10 Plastic kinematic 18.0 11 Plastic kinematic 8.0 12 Nonlinear elastic discrete beam 13 Rigid 14 Rigid 15 Rigid 16 Rigid 17 Rigid 18 Rigid 19 Rigid 20 Elastic 21 Elastic 22 Rigid

Sunil Kumar Sharma, Amity School of Engineering and Technology, Amity University Uttar Pradesh, Noida, India

Rakesh Chandmal Sharma, Maharishi Markandeshwar (Deemed to be University) Muliana, India

History

Received: 28 Nov 2017

Revised: 05 Jul 2018

Accepted: 30 Jul 2018

e-Available: 02 Nov 2018

doi:10.4271/02-11-04-0019

TABLE 1 Salient features of WAP-5 [23]. Feature WAP-5 Manufacturer Chittaranjan Locomotive Works, West Bengal, India Traction motor ABB's 6FXA 7059 three-phase squirrel cage induction motors (Torque 6930/10000 Nm, 1150 kW, 96% efficiency, 1583/3147 rpm) Train ratio 67:35:17 (three-stage gears) Axle load (tons) 19.5 Power Maximum: 4500 kW Tractive effort Maximum: 258 kN Diameter of wheel 1092 mm (new), 1016 mm (full worm) Wheelbase 13,000 mm Bogies Bo-Bo, Henschel Flexi float Body width 3.144 m Locomotive weight (tons) 78 Cab length 2434 mm Pantograph locked down 5437 mm height TABLE 2 Material property of WAP-5. Description IS-2062 Modulus of elasticity 210 GPa Tangent modulus 1 GPa Poisson's ratio 0.3 Density 7.85E-09 ton/cubic mm Yield stress 0.6 GPa Yield stress 250 MPa Ultimate stress 490 MPa TABLE 3 Meshing details of locomotive. Number of nodes 249,285 Number of element 150,367 Shell 98,950 Solid 21,586 Beam 28 TABLE 4 Mesh quality testing. Mesh metric Orthogonal quality Skewness quality Min 0.01 1.31E-10 Max 1.00 1.00 Average 0.89 0.21 Standard deviation 0.20 0.24 TABLE 5 Percentage elements and quality zones. S. No. Quality zone Orthogonal (%) Skewness (%) 1 Very good 69.23 78 2 Good 27.12 12 3 Acceptable 3.65 9 4 Unacceptable 0 1 TABLE 6 RND in the longitudinal direction. RND of right side RND of left side RND of central area Speed (km/hr) locomotive locomotive locomotive 25 110,542 111,398 1659 160 112,542 110,398 2756 200 91,542 91,998 3253 225 121,642 121,198 4445 Crash zone collapses Deformation at central Speed (km/hr) (ms) area (ms) 25 297 309 160 199 207 200 166 175 225 134 142

Printer friendly Cite/link Email Feedback | |

Author: | Sharma, Sunil Kumar; Sharma, Rakesh Chandmal |
---|---|

Publication: | SAE International Journal of Commercial Vehicles |

Article Type: | Report |

Date: | Oct 1, 2018 |

Words: | 3961 |

Previous Article: | Hydro-Pneumatic Energy Harvesting Suspension System Using a PSO Based PID Controller. |

Next Article: | Computational Fluid Dynamic Simulation of In-Cylinder Pressures to Validate High-Range VCR. |

Topics: |