Yuewei Ma, PhD Candidate – Delft University of TEchnology
Methodology for improvement of turnout performance via crossing shape optimisation
The presented methodology is based on the numerical/experimental analysis of the dynamic behaviour of the trainturnout system and a numerical optimisation technique[1, 2] Firstly, the dynamic interaction between the vehicle and turnout is analysed experimentally using the instrumented crossing (Figure 1). The measured data comprising of the 3D accelerations of the crossing nose and locations of the impact on the crossing nose (Figure 1) were collected on several turnouts. The measured data confirmed the numerical results presented earlier by the authors that the crossing nose geometry has significant influence on the dynamic forces in the crossing area.
Performance of the turnouts is also analysed numerically. The dynamic analysis of the vehicle/turnout and wheel/rail interaction is performed using numerical models on macro (multibody) and micro (explicit finite element) levels. The numerical models were validated using the collected geometry and acceleration measurement data. The rail geometry in the turnout crossing (1:15) comprising of the wing rail and crossing nose (Figure 2), is modelled here using Bsplines.
To optimise the shape of the crossing nose the parametrisation of the crossing geometry is proposed, using which the longitudinal height of the crossing nose and wing rail as well as the crosssectional shape of the nose rail can be varied. This parametrisation method is based on the manufacturing and maintenance process of the crossing, as it will be described during the presentation.
Turnout performance used as the criterion of the optimisation is represented by the combined objective function, which accounts for impact damage (the normal pressure S) and wear (energy dissipation in contact W) of the crossing nose.
Constrains are imposed on the values of the design variables to avoid unrealistic rail shapes, stability of the vehicle and location of the impact point on the crossing nose. Solution of such a problem is the set of compromised solutions (the Pareto) has found using the Multipoint Approximation Method.
Finally, one optimum design was chosen from the Pareto solutions and its robustness was checked by performing multiple simulations of the vehicle passing the turnout for various initial disturbances of the vehicle. The optimised crossing nose is implemented in the Dutch railway network and its performance assessed using the instrumented crossing device. One method to implement the improved geometry is proposed.
Conclusions and some recommendations on maintenance of the crossing nose geometry will be given.
 C. Wan, V. L. Markine, I. Y. Shevtsov, and “Improvement of vehicle–turnout interaction by optimising the shape of crossing nose,” Vehicle System Dynamics, 2014.
 V. L. Markine and I. Y. Shevtsov, “An Experimental Study on Crossing Nose Damage of Railway Turnouts in The Netherlands,” presented at the Proceedings of the Fourteenth International Conference on Civil, Structural and Environmental Engineering Computing (CC2013), 2013.
See full programme Wheel/Rail Interface & Switches