Nearly all aerospace structures contain curved panels in their design, including concept hypersonic vehicles. Curved panels are susceptible to dynamic snap-through when subjected to high amplitude loading. In order to properly model the statistical response of these structures to the random loading environment, the response must be computed over a long time window. This can be done using the full order nonlinear finite element (FE) model but is computationally expensive due to the large number of equations that must be solved at every time step. This is time consuming even on modern computing clusters, and this drastically limits design and analysis cycles of the vehicles. Nonlinear Reduced Order Models (NLROMs) have been developed recently which can compute the geometrically nonlinear response of thin panel structures far more efficiently. The accuracy of a NLROM can vary drastically depending on how it is created, specifically depending on which modes are included within the basis set and the strength of the static loads used to extract the nonlinear stiffness terms. This work proposes a procedure to check that an NLROM can reproduce certain static snap through responses. We show that if the NLROM accurately predicts the snap through behavior of the model to a distributed load, that it also tends to correctly reproduce the dynamic response. The load cases that are needed in terms of amplitude and modes included to assure that the NLROM reproduces the static snap through was examined and is presented.
展开▼
