Uncertainty Optimization of Vibration Characteristics in Engineering Machinery Powertrain Mounting Systems Using Monte Carlo and Genetic Algorithms

With the rising demand for high performance and reliability in engineering machinery, the vibration isolation performance and robustness of the powertrain mounting system (PMS) have become critical to overall machine performance. However, during service, rubber mounts are prone to environmental influences, causing significant stiffness deviations that render traditional optimization and analysis methods inadequate. To address this, this article proposes an uncertainty optimization strategy combining Monte Carlo and genetic algorithm (MC-GA), applied to design optimization accounting for stiffness uncertainty due to mount aging, to enhance vibration isolation robustness under large-scale stiffness fluctuations. The study first establishes a Monte Carlo analysis framework based on the statistical characteristics of retired mount stiffness and a dynamic model, systematically evaluating the impact of varying stiffness deviations on vibration characteristics under the original PMS configuration. On this basis, using initial mount stiffness as the optimization variable and considering extreme aging conditions, the MC-GA method is employed for uncertainty optimization of vibration characteristics. Real-vehicle test results demonstrate the method’s strong engineering applicability: in the scenario with a 50% increase in initial stiffness, the decoupling rate improved by an average of 4.2%, effectively reducing system vibration coupling; the vibration isolation rate from chassis to powertrain in the 20–80 Hz range during operation decreased by an average of 4.3 dB, effectively reducing the vibration transmitted to the powertrain.