Dynamic sensitivity-based finite element model updating for nonlinear structures using time-domain responses

Abstract

To obtain the numerical model for predicting the nonlinear dynamic responses with high accuracy, a dynamic sensitivity-based model updating approach by using the time-domain responses is proposed in this paper. The sensitivity analysis of time-domain response is derived by using the direct differentiation method. The objective function of the nonlinear model updating is constructed by minimizing the discrepancy between the measured and the calculated time-domain responses. The time-domain responses and the corresponding dynamic sensitivities are calculated synchronously. The repeated nonlinear dynamic analysis can be avoided to obtain dynamic sensitivity, which is independent of the perturbation step. Numerical examples of a Duffing-Van der Pol oscillator, a magnetometer boom, and a cantilever plate with multiple nonlinear supports are adopted to verify the method. Crucial issues about the measured noise and the selection of the targeted responses are also considered and discussed. The validation results show that the proposed method is effectively applied to model updating of nonlinear structure using time-domain response with good anti-noise performance, and the scheme for response points selection is reliable for guaranteeing the accuracy.