Kriging-PSO-based shape optimization for railway wheel profile

Abstract

The reduction of wheel-rail wear is a fundamental task in railway engineering that significantly affects the operating performance in the lifecycle. To improve the dynamic response and profile wear evolution performance of wheel-rail interaction, a shape optimization procedure for the railway wheel profile is proposed. First, the geometry modeling method, which ensures the continuity of first-order derivation of the wheel profile, is introduced to generate a large number of candidate profiles, and multibody dynamics simulation is conducted to analyze the dynamics response of the wheel profiles, including wear index, lateral force, lateral acceleration of the frame and derailment coefficient. Then, the Kriging model is constructed to establish the relationship between the design variables and objectives obtained by multibody dynamics simulation, and particle swarm optimization (PSO) is employed to evaluate the optimal parameters for wheel profile that simultaneously considers wheel wear, stability, and lateral force. Finally, the performance of the wheel-rail interaction is evaluated to demonstrate the effectiveness of the proposed method. The numerical simulation result indicates that the optimized wheel profile not only has good performance, including contact state, pressure, and friction at the design stage, but also the physical performance is acceptable after a long-term profile evolution during service, which the maximum wear depth of the optimal wheel profile averagely decreases over 10 % in long-term wear evolution.