Identification of time‐delay Markov jump autoregressive exogenous systems with expectation‐maximization algorithm

Abstract

This paper is concerned with the identification problem of the Markov jump autoregressive exogenous system with an unknown time delay. The considered problem is solved using the expectation-maximization algorithm, which estimates the parameters of local models, Markov transition probabilities, and time delay simultaneously. A numerical example and a simulated continuous fermentation reactor example are given to illustrate the capability of the proposed method. It shows that the influences of time delay during identification can be overcome by the proposed algorithm effectively. KEYWORDSEM algorithm, Markov jump system, parameter identification, unknown constant time delay 1920

Parameter estimation for time‐delay systems based on the frequency responses and harmonic balance methods

“…Using the simulation model parameters and the input signal generates the output signal {y(t k ), k = 1, 2, … }. First apply the SG algorithm in (11) to (15) to estimate the parameters C i and D i (i = 1, 2, 3) of the output response, respectively. v i (t) is the measurement error (ie, noise) for the output y i (t).…”

“…10 Chen et al solved the identification problems of the Markov jump autoregressive exogenous systems with an unknown time delay using the expectation maximization algorithm. 11 The classical parameter estimation methods, including the impulse response method, the step response method, and the frequency response method, are often used for system modeling. Some parameter estimation algorithms are proposed to identify the parameters of the transfer functions of the linear time-invariant systems.…”

Online identification of time‐delay jump Markov autoregressive exogenous systems with recursive expectation‐maximization algorithm

“…For the identification problem of time‐delay jump Markov autoregressive exogenous (JMARX) systems, a great difficulty exists due to the coupling of the time‐delay, mode, and parameters. The BEM algorithm has been employed to solve this problem in Reference . However, to the best knowledge of the authors, there is no report up to now dealing with the identification issue of the JMS with time‐delay in a recursive manner.…”

Variational Bayesian identification for bilinear state space models with Markov‐switching time delays

“…System identification finds a mathematical model to approximate the system's dynamics by using the collected information 1,2 and has been applied to various areas. [3][4][5] In recent years, parameter identification for nonlinear systems has been an active topic. 6 As a typical class of nonlinear system, the bilinear model has an obvious adaptive structure, 7 which makes it show good performance in describing some complex industrial processes and plants, such as hydraulic systems, gas stoves, heat exchange systems, and so on.…”