Wheel profile optimization of high-speed locomotive to improve railway line adaptability

Aiming at the low-frequency carbody swaying phenomenon of the locomotive belonging to the power concentrated electric multiple unit (EMU) for high-speed passenger transport, which operated in the early stage after the wheel profile is repaired as the JM₃-32 thin flange profile according to the standard, the profile is optimized to improve the locomotive’s dynamic performance. The optimization objective is to improve the equivalent conicity and its distribution concentration of the wheel profile under various rail profiles and rail cants conditions, as well as to improve its wear performance. The mapping relationship between the design variables and the objective functions is realized by constructing the radial basis function neural network (RBFNN) surrogate model. The Nondominated Sorting Genetic Algorithm II (NSGA-II) is used to optimize the position and size of the four arcs in the rolling circle contact area of the wheel profile. The wheel-rail contact characteristics and vehicle dynamics simulation of the wheel profile before and after optimization are compared to reveal the optimization effects. The results show that under different wheel-rail matching conditions, the equivalent conicity of the optimized wheel profile is significantly improved compared with the original wheel profile, and the dispersion degree of its distribution is significantly reduced at the same time. The nominal equivalent conicities corresponding to the 3 mm lateral displacement of the wheelset are all about 0.1. The nonlinear critical velocity is increased by more than 290 km/h under the condition of 1/20 rail cant. Under the four wheel-rail matching conditions, the Locomotive will not have the problem of lateral ride comfort deterioration and vehicle swaying when running at high speed. The curve passing performance and wear performance are also improved.