Robust and reliable modification design of traction arc cylindrical gears for high-speed electric multiple units

To enhance the smoothness, quietness, and reliability of traction gear transmissions in high-speed electric multiple units operating under uncertain conditions, this study introduces an innovative arc cylindrical gear system along with a robust optimization strategy for gear modification. Center distance error and material elastic modulus are selected as key disturbance factors, and their uncertainty distributions are modeled using fuzzy interval theory and the optimal level-cut method. A robust optimization model is developed with the maximum profile modification amount, modification length, and arc tooth line radius as design variables to analyze the correlation between modification parameters and noise generation. The model aims to minimize radiated noise while satisfying noise reliability constraints, and the optimal robust modification scheme is obtained using a genetic algorithm. The results indicate that the optimized gear design achieves a 63.7% reduction in transmission error, a 48.4% decrease in maximum contact stress at the meshing-out position, and a 16.3% drop in radiated noise, while maintaining a robust reliability index R_s of 98.02% across the full range of disturbance parameters. This study offers theoretical guidance and practical strategies for achieving robust and reliable design of traction gears in next-generation high-speed trains.