2-Copula and Simple Criterion based Algorithm for Identification of Time Delay Dynamic Systems

In this paper, the identification of time-delay dynamic systems is considered. Due to the limitation of the hardware and other reasons, time-delay systems exist widely in the industrial field. The identification of this kind of system includes time delay identification and parameter identification. Firstly, a time delay identification method based on the 2-copula is studied, which can obtain an accurate estimation of time delay when the parameters are unknown. Secondly, a compact multi-innovation stochastic gradient algorithm named multi-gradient algorithm is proposed. The algorithm uses the simple criterion of selecting step size in the stochastic gradient method to determine the stacking length and obtains the available stacking length based on the bisection search method. A numerical example shows the effectiveness of the algorithm.