Abstract: In this paper we consider time scale synchronization between two different timedelay systems. Due to existence of intrinsic multiple characteristic time scales in the chaotic time series, the usual definition for the calculation of phase failed. To define the phase, we have used empirical mode decomposition and the results are compared with those from continuous wavelet transform. We ...

In this study, we present a general form of nonlinear two-time-scale systems, where singular perturbation analysis is used to separate the dynamics slow and fast subsystems. Machine learning techniques are utilized approximate both Specifically, recurrent neural network (RNN) feedforward (FNN) predict state vectors, respectively. Moreover, investigate generalization error bounds for these machi...