A study of the modal strain energy method for viscoelastically damped structures

Abstract The modal strain energy (MSE) method has been proposed to estimate the modal loss factors or modal damping ratios of structures with viscoelastic dampers. There are certain assumptions made in deriving the MSE method, such as the computation of the modal loss factor directly from the ratios of the imaginary and real parts of the eigenvalues, and the neglect of the influence of the imaginary mode shapes, etc. These assumptions may result in overestimating the modal damping ratios of viscoelastically damped structures when the added damping is high. In this study, the effect of the assumptions made by the MSE method is investigated, and modified formulations of the MSE method are derived. The modified MSE method removes the assumptions made in the original MSE method. Furthermore, earthquake responses of a complex stiffness system and a linear viscous damping system, of which the modal damping ratios are estimated by the MSE method, are compared. Study results indicate that the difference arising from the assumptions becomes significant when the damping ratio is larger than 20%. For the illustrated non‐proportionally damped system, the effect of imaginary mode shapes on the MSE method can be neglected when the damping ratio is smaller than 20%. It is then concluded that, for most engineering applications with design damping ratios smaller than 20%, the conventional linear viscous damping model with the MSE method may result in solutions in good agreement with those obtained by the more rigorous complex stiffness model.