Optimization of Crossing Geometry in Railway Turnouts

Transcription

1 Optimization of Crossing Geometry in Railway Turnouts Dipl.-Ing. Dirk Nicklisch DB Netz AG Technology (I.NVT 8) D-8939 Munich, Germany SIMPACK User Meeting Salzburg

2 Overview Contents 1. Introduction 2. Simulation model and model validation 3. Investigation of alternative crossing designs 4. Simulation of material degradation in service 5. Conclusion and future work DB Netz AG, Dirk Nicklisch, I.NVT 8 2

3 Introduction switch panel crossing panel crossing nose with wing rails facing move point of transition RCF due to severe impact loads vertical wheel movement vertical contact force Photo: DB AG (A. Zoll) Damage on running surface of crossing nose wheel passing a crossing DB Netz AG, Dirk Nicklisch, I.NVT 8 3

4 Overview Contents 1. Introduction 2. Simulation model and model validation< 3. Investigation of alternative crossing designs 4. Simulation of material degradation in service 5. Conclusion and future work DB Netz AG, Dirk Nicklisch, I.NVT 8 4

5 SIMPACK simulation model facing move, running direction crossing nose wing rail stock rail closure rail Y Q flexible rails (Timoshenko beams) Q Y Elastic track model (FEM) implemented in SIMPACK (user routine of DB Netz AG) flexible sleeper (Timoshenko beam) DB Netz AG, Dirk Nicklisch, I.NVT 8 5

6 Model validation Comparison of measured and simulated vertical contact forces of two passing wheels T T 1 R 1 R 2 T 3 T 2 direction of traffic Positions of strain gauges (x) (T i shear forces, R j rail seat forces) Vertical wheel-rail contact force [kn] Wheel 1 - sim Wheel 2 - sim Wheel 1 - meas Wheel 2 - meas Summation: Q = T + R 1 + R 2 + T Time [s] DB Netz AG, Dirk Nicklisch, I.NVT 8 6

7 Overview Contents 1. Introduction 2. Simulation model and model validation 3. Investigation of alternative crossing designs 4. Simulation of material degradation in service 5. Conclusion and future work DB Netz AG, Dirk Nicklisch, I.NVT 8 7

8 Optimization of geometry Crossing geometry Kinked Ramp for EH 6-5-1:12 (Trajectory of profiling tool) x 3 Theoretical wheel transition point Conventional Kinked Ramp x z z Tip of crossing nose All measures given in mm. Shape of profiling tool DB Netz AG, Dirk Nicklisch, I.NVT 8 8

9 Optimization of geometry Crossing geometry Herzstückgeometrie MaKüDe EH 6-5-1:12 for MaKüDe EH 6-5-1:12 Wheel profile S12 DB Netz AG, Dirk Nicklisch, I.NVT 8 9

10 Vertical wheel movement EH6 5 1:12 Comparison of vertical wheel movement (profile S12) Comparison of vertical wheel movement (quasi-static) 1.5 Begin of wing rail End of crossing nose Vertical displacement [mm] Facing move Nose tip -3.5 Wheel transition Distance [m] Kinked ramp (S12) Kinked ramp (worn) Kinked ramp (short) Wing rail rail MaKüDe (S12) Kinked ramp (S12) Kinked ramp (short) Wing rail MaKüDe (S12) Current standard (S12) Wing rail rail MaKüDe (worn) Current standard (S12) Current standard (worn) Current standard (hollow worn) DB Netz AG, Dirk Nicklisch, I.NVT 8 1

11 Evaluation of Contact Stresses MBS simulation 3D Stress calculation Creepages, Friction coeff., Penetration in contact patch CONTACT (Prof. J. J. Kalker) Normalspannung Tangentialspannung p [MPa] tang[mpa] y mm x mm 1-5 y mm x mm 1 pmax(16,16)i(.5,.3)mm = 1848.[MPa] svmax(1,24)i( -6.2, 5.7)mm = 196.[MPa] shmax(16,16)i(.5,.3)mm = 2793.[MPa] tamax(1,24)i( -6.2, 5.7)mm = 1252.[MPa] Reibleistung Durchdringung 15 Tangentialspannung N = 333.3[kN] TT = 41.3[kN] Tx =.8[kN] Ty = -41.3[kN] 3 1 DB Netz AG, Dirk Nicklisch, I.NVT 8 11 pw [MPa] y mm x mm 1 x mm y mm

12 Equivalent Contact Stress Calculation of equivalent contact stresses according to v. Mises σ e = 1 2 xx yy yy zz zz xx xy yz zx 2 [ ] ( σ σ ) + ( σ σ ) + ( σ σ ) + 3 ( τ + τ + τ ) DB Netz AG, Dirk Nicklisch, I.NVT 8 12 (Figures from PhD thesis of Dr. André Theiler, IfB Berlin [1])

13 Discussion of simulation results Normal forces and equivalent stresses in relation to Standard geometry DB Netz AG, Dirk Nicklisch, I.NVT 8 13

14 Discussion of simulation results Time histories of Normal force and von Mises stress 1 2 T x T y N Normal force Equivalent stress Normal force [kn] von Mises stress [MPa] 1 1 T x T y N 5 5 Running direction (facing move) Distance [m] DB Netz AG, Dirk Nicklisch, I.NVT 8 14

15 Overview Contents 1. Introduction 2. Simulation model and model validation 3. Investigation of alternative crossing designs 4. Simulation of material degradation in service 5. Conclusion and future work DB Netz AG, Dirk Nicklisch, I.NVT 8 15

16 CHALMERS CHARMEC Simulation methodology I F v II Simulation of dynamics 3D elasto-plastic contact simulations IV III Summation and smoothing of total profile change a) Plasticity calculation b) Wear calculation DB Netz AG, Dirk Nicklisch, I.NVT 8 16 Source: CHARMEC, see [2]

17 CHALMERS CHARMEC Parameter variation Simulation Schedule Facing move on through route Stochastic vehicle parameters: Wheel profiles - Six different wheel profiles - Three lateral wheel positions - Three vehicle speeds: 9, 12, 16 km/h - Two vehicle types: BR11 and BR = 18 possible combinations Creation of a sample set based on estimated probability distributions Initial lateral wheel positions on crossing entry Reduction of sample size by Latin Hypercube Sampling DB Netz AG, Dirk Nicklisch, I.NVT 8 17 centred close far

18 CHALMERS CHARMEC Simulation results Comparison with field measurements Degradation rate is larger during the first weeks after installation and then it seems to stabilise Comparison of measured and simulated profiles after five weeks of traffic Good qualitative agreement - Largest degradations found for similar longitudinal positions along crossing nose - Same order of degradation magnitude DB Netz AG, Dirk Nicklisch, I.NVT 8 18

19 Overview Contents 1. Introduction 2. Simulation model and model validation 3. Investigation of alternative crossing designs 4. Simulation of material degradation in service 5. Conclusion and future work DB Netz AG, Dirk Nicklisch, I.NVT 8 19

20 Conclusions Conclusions Optimization of crossing geometry is essential to reduce the dynamic loads and consequently the Life Cycle Cost of railway turnouts. The simulation model developed by DB has been successfully validated by wayside force measurements on an in-service crossing of type EH 6-5-1:12. When comparing different crossing designs the maximum equivalent stresses calculated with CONTACT may show different trends than the maximum normal forces as the locations of the maximum stresses may differ significantly from those of the maximum forces. For relevant conclusions regarding material degradation the elasto-plastic material behavior has to be taken into account using calibrated non-linear material models. The related simulation methodology developed in INNOTRACK has been demonstrated on a crossing of R35HT steel showing good qualitative agreement between simulation and measurement after five weeks of train traffic. DB Netz AG, Dirk Nicklisch, I.NVT 8 2

21 Future work Remaining tasks Validation of simulation methodology for a longer period of operation (done for 5 weeks of traffic only) Investigation of 3 different crossing materials regarding plastic deformation and wear Further improvement of crossing geometry MaKüDe Comparison of long-term degradation of innovative crossing geometries DB Netz AG, Dirk Nicklisch, I.NVT 8 21

22 Thank you for listening! The author wishes to acknowledge the support of the European Commission under the FP6 project INNOTRACK and would like to thank all the project partners for the constructive and fruitful cooperation. [1] A. Theiler: Numerische Untersuchung der Beanspruchung im Rad-Schiene-Kontakt, Dissertation TU Berlin, 25 [2] D. Nicklisch, J. C. O. Nielsen, M. Ekh, A. Johansson, B. Pålsson, J. M. Reinecke, A. Zoll: Simulation of wheel rail contact forces and subsequent material degradation in switches & crossings, Proceedings 21 st International Symposium on Dynamics of Vehicles on Roads and Tracks, Stockholm, 29 DB Netz AG, Dirk Nicklisch, I.NVT 8 22