Title: Parameter identification of fractional order CARMA model based on least squares principle

Authors: Jiali Rui; Junhong Li

Addresses: School of Electrical Engineering, Nantong University, Nantong, Jiangsu, China ' School of Electrical Engineering, Nantong University, Nantong, Jiangsu, China

Abstract: The fractional order model is more accurate than the integer order model when describing the actual system. This paper studies the fractional order CARMA model identification, and derives the identification expression of the fractional order CARMA model through the definition of Grünwald-Letnikov fractional order differentiation. In order to identify the unknown parameters of the model with coloured noise, the least squares based iterative identification algorithm and the recursive extended least squares algorithm are respectively derived based on the least squares principle. Then two simulation examples are given. The simulation results show that the errors of the parameter estimation obtained by the two algorithms are small, which proves the effectiveness of the proposed algorithms.

Keywords: fractional order model; parameter estimation; system identification; least squares; coloured noise.

DOI: 10.1504/IJCAT.2022.126096

International Journal of Computer Applications in Technology, 2022 Vol.69 No.1, pp.25 - 35

Received: 27 Jun 2021
Accepted: 22 Aug 2021
Published online: 11 Oct 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article