Identification of nonlinear dynamical systems with time delay

Abstract This paper proposes a technique to identify nonlinear dynamical systems with time delay. The sparse optimization algorithm is extended proposed combines cross-validation techniques from machine learning for automatic model selection and an algebraic operation preprocessing signals filter the noise removing dependence on initial conditions. We further integrate bootstrapping resampling regression obtain statistical properties of estimation. use Taylor expansion parameterize in this computationally efficient robust noise. A Duffing oscillator simulated demonstrate efficiency accuracy technique. An experimental example rotary flexible joint presented validate method.