Integrated shape-morphing and metamodel-based optimization of railway wheel web considering thermo-mechanical loads

While the wheel and the rail are in contact, the stress on the wheel affects the safety of the vehicle and causes ride discomfort and noise. Therefore, the design of the wheel to reduce the damage is needed. To solve these problems, the shape optimization on the railway wheel web was carried out using the finite-element analysis and shape-morphing technique. First, thermo-mechanical analysis, with consideration to mechanical and thermal loads under tread braking, was developed and the contact pressure and stress generated on the wheel were obtained. Then, the integrated shape-morphing design process was proposed to improve the efficiency of the shape optimization. It made it possible to parameterize the finite-element models and modify them directly without returning to the computer-aided design model. Based on the analysis results, we performed the metamodel and genetic algorithm based optimization by using response surface method, Kriging and Gaussian process, then compared their optimal solutions. Optimization proceeded in two stages, two-dimensional optimization under axisymmetric conditions and three-dimensional optimization in which the shape changed in the period of 60°. For the post-optimization work, fatigue evaluation for the verifications on initial and optimal wheel models was performed and the results were discussed.