Title: Several structure pool based identification algorithms for ARX models: order and parameter estimation
Authors: Jianwei Lu
Addresses: School of Information Engineering, Changzhou Institute of Industry Technology, Changzhou, 213164, China
Abstract: Several structure pool based identification algorithms are proposed for autoregressive exogenous (ARX) models with unknown order in this study. Since the order of the ARX model is unknown, a structure pool which contains various different information products is provided, and then the gradient iterative and two-direction stochastic gradient algorithms are provided to estimate the order and the unknown parameters simultaneously. The proposed algorithms can be applied for systems with unknown orders and parameters, thus are more promising in engineering practice. The simulation example is utilised to validate the efficiency of the proposed algorithms.
Keywords: order identification; gradient iterative algorithm; ARX model; two-direction stochastic gradient algorithm; parameter estimation; structure pool.
International Journal of Modelling, Identification and Control, 2020 Vol.36 No.3, pp.233 - 239
Received: 12 Aug 2020
Accepted: 07 Oct 2020
Published online: 02 Aug 2021 *