The Robustness of an Efficient Probabilistic Data-Based Tool for Simulating the Nonstationary Response of Nonlinear Systems

The stochastic response of dynamic systems has been an area of considerable interest for some time in the analysis of risk and structural reliability. The authors, in previous work, have developed a method which can analyze the response of linear multi-degree-of-freedom systems to completely general data-based nonstationary excitations in a highly efficient and analytical form. The authors extended this work to nonlinear system response by using equivalent linearization techniques. This paper explores the range of application of the extension to the analysis of nonlinear systems through the use of real and simulated data-sets. In particular, sensitivity issues of non-Gaussianity of the excitation data, and the degree of response nonlinearity of the dynamic system, are investigated for their effect on the estimated response using this equivalent linearization based approach.