Perturbing Structural Design Towards Minimizing Variation in Vibratory Response

Real structures are always subject to uncertainties due to material imperfection, machining tolerance, and assemblage error, etc. These uncertainties lead to variations in structural vibratory responses. In order to reduce the likelihood of unexpected failures in structures, we need to minimize the response variations, which is the underlying idea of robust design. In this paper, we present an inverse sensitivity-based algorithm that allows us to tailor the structural design such that, under the same level of uncertainties, the response variations can be effectively reduced. We first develop a direct relation between the structural uncertainties and the response variations including the means and variances. We then formulate an optimal identification algorithm that will yield design perturbation to minimize the response variances while maintaining the mean values. Case analyses are carried out to validate the validity and efficiency of the new algorithm.