Optimisation of Railway Switches and Crossings

Methods for simulation-based optimisation of the design of railway turnouts (switches & crossings, S&C) are developed and demonstrated. Building on knowledge of dynamic wheel–rail interaction in turnouts, it is investigated how rail profile degradation can be reduced by the optimisation of geometry and component stiffness of the track superstructure. It is assumed that reduced rail profile degradation will reduce the Life Cycle Cost (LCC) of turnouts. In order to obtain robust optimised designs that perform well in situ, the influence of spread in traffic parameters, such as wheel profile and wheel–rail friction coefficient, is accounted for in the optimisations. For this purpose, studies of the correlation between wheel profile characteristics and damage in S&C are performed to allow for an efficient parameter sampling using the Latin Hypercube Sampling method. Track gauge optimisation in the switch panel is performed using a multi-objective optimisation approach to highlight the design trade-off in performance between different traffic routes and moves. The objective is to minimise rail and wheel wear as estimated by the energy dissipation in the wheel‒rail contacts. As track gauge widening affects the switch rail design, the switch rail geometry is linked to the gauge widening in the parameterisation. It is found that gauge configurations with a large maximum gauge widening for the straight stock rail are optimal for both the through and diverging routes, while the results for the curved stock rail show a more significant route dependence. A method for the optimisation of switch rail profile geometry is presented, where the geometry parameterisation is inspired by a manufacturing process for switch rails. It is found that increased profile height and increased profile shoulder protuberance are preferred to reduce the energy dissipation and wheel‒rail contact pressures when a nominal S1002 wheel profile is used as input. It is concluded that accurate constraints on allowable switch rail loading need to be established to determine the feasible design space for switch rail geometry optimisation. A method for the optimisation of crossing geometry is also introduced. The rail cross-sections are optimised for minimum wheel‒rail contact pressure. Further, the longitudinal height profiles of the wing rails and crossing nose are optimised to minimise an estimate of the accumulated damage in the transition zone. The optimisation is computationally efficient which makes it possible to account for very large samples of wheel profiles. An investigation and demonstration of the constraints imposed on the crossing design by the spread in profile and lateral displacement of passing wheels is presented. Supplementary to the optimisation studies is the comparison of simulation results to field measurement data to evaluate and validate the accuracy of the utilised model of dynamic vehicle‒track interaction, as well as a demonstration of a methodology that simulates rail profile degradation for a given mixed traffic situation.

[1]  V L Markine,et al.  Optimization of a wheel profile accounting for design robustness , 2011 .

[2]  Aaas News,et al.  Book Reviews , 1893, Buffalo Medical and Surgical Journal.

[3]  Joseph G. Pigeon,et al.  Statistics for Experimenters: Design, Innovation and Discovery , 2006, Technometrics.

[4]  V. Markine,et al.  Combatting RCF on switch points by tuning elastic track properties , 2011 .

[5]  D Lyon,et al.  THE EFFECT OF TRACK AND VEHICLE PARAMETERS ON WHEEL/RAIL VERTICAL DYNAMIC FORCES , 1974 .

[6]  Hiroyuki Sugiyama,et al.  Railroad Vehicle Dynamics: A Computational Approach , 2007 .

[7]  T. G. Pearce,et al.  Prediction of wheel profile wear , 1990 .

[8]  Elena Kabo,et al.  An engineering model for prediction of rolling contact fatigue of railway wheels , 2002 .

[9]  Jens C. O. Nielsen,et al.  Dynamic interaction between train and railway turnout: full-scale field test and validation of simulation models , 2008 .

[10]  Asier Alonso,et al.  Rail vehicle passing through a turnout: analysis of different turnout designs and wheel profiles , 2012 .

[11]  Mattias Wahde,et al.  Biologically inspired optimization methods , 2008 .

[12]  Jens C. O. Nielsen,et al.  Stochastic analysis of dynamic interaction between train and railway turnout , 2008 .

[13]  Jens C. O. Nielsen,et al.  Geometry and stiffness optimization for switches and crossings, and simulation of material degradation , 2010 .

[14]  C. Esveld Modern railway track , 1989 .

[15]  Simon Iwnicki,et al.  Handbook of railway vehicle dynamics , 2006 .

[16]  Gang Xie,et al.  A method to determine the two-point contact zone and transfer of wheel–rail forces in a turnout , 2010 .

[17]  Paul Allen,et al.  Optimisation of track stiffness, track gauge and rail inclination along a turnout using a multibody simulation tool , 2009 .

[18]  Sebastian Stichel,et al.  Influence of switches and crossings on wheel profile evolution in freight vehicles , 2013 .

[19]  Elias Kassa Dynamic Train-Turnout Interaction – Mathematical Modelling, Numerical Simulation and Field Testing , 2007 .

[20]  V. L. Markine,et al.  An experimental study on crossing nose damage of railway turnouts in The Netherlands , 2013 .

[21]  Arne Nissen Development of life cycle cost model and analyses for railway switches and crossings , 2009 .

[22]  Jens Reinecke,et al.  Simulation of wheel–rail contact and subsequent material degradation in switches & crossings , 2009 .

[23]  M. R. Bugarín,et al.  Improvements in railway switches , 2002 .

[24]  Ahmed A. Shabana,et al.  Dynamics of Multibody Systems , 2020 .

[25]  Dirk Nicklisch,et al.  Recommendation of, and scientific basis for, optimisation of switches a crossings : part 1 , 2009 .

[26]  C. S. Bonaventura,et al.  Switch and crossing inspections and maintenance management using handheld computers , 2008 .

[27]  K. Knothe,et al.  Rail Vehicle Dynamics , 2016 .

[28]  A. Olsson,et al.  On Latin hypercube sampling for structural reliability analysis , 2003 .

[29]  Jeremy Evans,et al.  Vehicle/track interaction and rolling contact fatigue in rails in the UK , 2006 .

[30]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[31]  Rolf Dollevoet,et al.  Improvement of train-track interaction in turnouts by optimising the shape of crossing nose , 2013 .

[32]  Arne Nissen Analys av statistik om spårväxlars underhållsbehov , 2005 .

[33]  J. J. Kalker,et al.  A Fast Algorithm for the Simplified Theory of Rolling Contact , 1982 .

[34]  J. Barbera,et al.  Contact mechanics , 1999 .

[35]  R. K. Ursem Multi-objective Optimization using Evolutionary Algorithms , 2009 .