In this paper we propose a new technique for obtaining an approximate
probability density for the response of a general non-linear system un
der Gaussian white noise excitations. In this new technique, the origi
nal non-linear system is replaced by another equivalent non-linear sys
tem structured by a polynomial formula, for which the exact solution o
f a stationary probability density function is obtainable. Since the e
quivalent non-linear system structured in this paper originates direct
ly from certain classes of real non-linear mechanical systems, we will
apply this technique to some very challenging non-linear systems in o
rder to show its power and efficiency. The result of the calculations
shows that applying the techniques presented here can yield the exact
stationary solutions for non-linear oscillators. This is obtained by u
sing an energy-dependent system, and for a non-linearity of a more com
plex type. A more accurate approximate solution is then available, and
is compared with the approximation. Application of the technique is i
llustrated by examples. Copyright (C) 1996 Elsevier Science Ltd.