Identification of Hammerstein-Wiener models using Hamiltonian Monte Carlo

Title: Identification of Hammerstein-Wiener models using Hamiltonian Monte Carlo
Creator: Holdsworth, James R. Z.; Wills, Adrian G.
Relation: 20th IFAC Symposium on System Identification SYSID 2024. Proceedings of the 20th IFAC Symposium on System Identification SYSID 2024 [presented in IFAC-PapersOnLine, Vol. 58 No. 15] (Boston, United States 17-19 July, 2024) p. 456-461
Publisher Link: http://dx.doi.org/10.1016/j.ifacol.2024.08.571
Publisher: Elsevier
Resource Type: conference paper
Date: 2024
Description: This paper develops and illustrates a process for the identification of Hammerstein–Wiener model structures using Hamiltonian Monte Carlo. A central aspect is that a very general situation is considered wherein multi-variable data, non-invertible Hammerstein and Wiener nonlinearities, and Gaussian stochastic disturbances both before and after the Wiener non-linearity are all catered for. The posterior distribution of the parameters is directly sampled from and these samples can then be used to quantify the uncertainty of the system model estimates, and to calculate expected values of various quantities.
Subject: nonlinear system identification; Bayesian Methods; uncertainty quantification; Hamiltonian Monte Carlo
Identifier: http://hdl.handle.net/1959.13/1517423
Identifier: uon:57107
Language: eng
Reviewed

		Thumbnail	File	Description	Size	Format