Parameter estimation for chaotic systems with and without noise using differential evolution-based method

We present an approach in which the differential evolution (DE) algorithm is used to address identification problems in chaotic systems with or without delay terms. Unlike existing considerations, the scheme is able to simultaneously extract (i) the commonly considered parameters, (ii) the delay, and (iii) the initial state. The main goal is to present and verify the robustness against the common white Guassian noise of the DE-based method. Results of the time-delay logistic system, the Mackey-Glass system and the Lorenz system are also presented.