Accelerated dynamics model updating method based on multigrid reduced-order technique

Discrepancies between finite element models and the dynamic characteristics of real structures are common and often stem from various uncertainties and assumptions. These include simplifications, manufacturing inaccuracies, and material approximations. To enhance the accuracy of dynamic model predictions, finite element models are typically updated using experimental data. This paper introduces an accelerated dynamics model updating method based on a multigrid reduced-order technique. A comprehensive sensitivity analysis is conducted to classify model parameters, thereby reducing the number of parameters involved in the optimization process. By integrating experimental data from modal analysis and frequency response functions (FRFs), the finite element model is updated to better reflect the actual dynamic behavior. Additionally, the multigrid reduced-order technique significantly enhances the computational efficiency of the FRF during the optimization procedure. The proposed model updating approach is validated through simulations of the GARTEUR aircraft model and cantilever vibration experiments. The results demonstrate that the proposed method effectively improves the accuracy of the dynamic model, achieving an efficiency improvement of approximately two orders of magnitude compared to traditional full-order frequency response-based dynamic model updating approaches.