Nonstationary probability densities of system response of strongly nonlinear single-degree-of-freedom system subject to modulated white noise excitation (Q656907)

The authors consider a strongly nonlinear oscillator under modulated white noise excitation, described by the following equation of motion \[ \ddot x(t)+\varepsilon f(x(t),\dot x(t))\dot x(t)+ g(x(t))=\sqrt{\varepsilon}\sigma(t)\dot w(t),\tag{1} \] where \(x(t)\) and \(\dot x(t)\) are the displacement and the velocity, respectively, \(\varepsilon\) is a small positive parameter, \(f(x,\dot x)\) and \(g(x)\) are nonlinear functions, such that \(g(x)= -g(-x)\), \(\dot w(t)\) is a Gaussian white noise with intensity 2D, and \(\sqrt{\varepsilon}\sigma(t)\geq 0\) is a deterministic modulating function. To obtain the nonstationary probability densities of state response and of amplitude response together with the statistics moments of amplitude response, the authors use the standard way, i.e. they apply the stochastic averaging approach. Next, after the derivation of averaged Fokker-Planck-Kolmogorov equation for the amplitude, they find an approximate solution in the form of series expansion. To illustrate the obtained results, the authors consider a van der Pol-Duffing oscillator under modulated Gaussian white noise. A numerical comparison with Monte Carlo simulations is given.