An Improved Dynamic Model Updating Method for Multistage Gearbox Based on Surrogate Model and Sensitivity Analysis

Model updating can improve the accuracy of the multistage gearbox dynamic model (MGDM). However, owing to the time-consuming calculations and numerous parameters of the MGDM, previous updating methods are not suitable for updating the MGDM. To solve the above problems, an improved dynamic model updating method that is suitable for updating the MGDM is proposed in this paper. Compared with previous methods, this updating method presents two improvements: (1) In the iterative calculation process, the surrogate model is used to replace the time-consuming MGDM, thereby solving the problem of low computational efficiency; (2) before the updating calculation, Sobol sensitivity analysis is performed to screen out updating parameters from numerous parameters. This study shows that the accuracy of the updated MGDM improved significantly compared with the initial model, thereby enabling a more accurate prediction of the dynamic characteristics of the actual system. This proves the effectiveness of the improved updating method.