| Abstract: | The rate of convergence of the conventional mode superposition method can be significantly improved for systems subjected to relatively low frequency loadings by adding static correction terms to approximate the contribution of higher modes neglected by the use of a truncated eigenbasis. Two computational variants of the classical mode-acceleration method are developed for that purpose. It is recommended that the pseudo-static displacement correction vector should be computed from an expansion of the flexibility matrix in terms of the retained eigenvectors for actual computer implementation. |
