Data driven forecasting of aperiodic motions of non-autonomous systems

In the present effort, a data-driven modeling approach is undertaken to forecast aperiodic responses of non-autonomous systems. As representative system, harmonically forced Duffing oscillator considered. Along with it, an experimental prototype studied. Data corresponding chaotic motions are obtained through simulations oscillators hardening and softening characteristics experiments bistable oscillator. Portions these datasets used train neural machine make response predictions forecasts for on attractors. The constructed by using deep recurrent network architecture. conducted different numerical time-series data confirm effectiveness forecasting system responses.