Optimization Method for Powertrain Mounting Systems Based on Enhanced Genetic Algorithms

A methodology for optimizing natural properties of a powertrain for an electric vehicle has been presented. A model with six-degree-of-freedom was proposed utilizing ADAMS, and the natural frequencies and energy distribution of the powertrain are estimated using the proposed model. The calculated natural frequencies and energy distribution shown that the initial design of mount stiffness does not meet requirements of natural frequency and decoupling ratio, and vibration isolation standards. To overcome the limitations of conventional optimization techniques, a non-dominated sorting genetic algorithm (NSGA) was adopted for the enhancement optimization the mounts parameters. The optimization objectives included the refinement of the decoupling rates and frequency distribution at all mounting directions. Stiffness parameters of the mounts were optimized via the NSGA. The optimized results confirmed significant improvements for powertrain natural characteristics. This study presented an effective optimization approach for design of electric vehicle powertrain mounting systems.