Identification of fractional-order transfer functions using exponentially modulated signals with arbitrary excitation waveforms

Galvão, R. K. H., Tiexeira, M. C. M., Assunção, E., Paiva, H. M. and Hadjiloucas, S. ORCID: https://orcid.org/0000-0003-2380-6114 (2020) Identification of fractional-order transfer functions using exponentially modulated signals with arbitrary excitation waveforms. ISA Transactions, 103. pp. 10-18. ISSN 0019-0578 doi: 10.1016/j.isatra.2020.03.027

Abstract/Summary

This paper proposes a new identification method based on an exponential modulation scheme for the determination of the coefficients and exponents of a fractional-order transfer function. The proposed approach has a broader scope of application compared to a previous method based on step response data, in that it allows for the use of arbitrary input signals. Moreover, it dispenses with the need for repeated simulations during the search for the best fractional exponents, which significantly reduces the computational workload involved in the identification process. Two examples involving measurement noise at the observed system output are presented to illustrate the effectiveness of the proposed method when compared to a conventional output-error optimization approach based on the polytope algorithm. In both examples, the proposed method is found to provide a better trade-off between computational workload and accuracy of the parameter estimates for different realizations of the noise. Keywords: Fractional order systems, system identification, transfer functions

Item Type	Article
URI	https://reading-clone.eprints-hosting.org/id/eprint/89796
Item Type	Article
Refereed	Yes
Divisions	Interdisciplinary Research Centres (IDRCs) > Centre for Integrative Neuroscience and Neurodynamics (CINN) Life Sciences > School of Biological Sciences > Department of Bio-Engineering
Publisher	Elsevier
Download/View statistics	View download statistics for this item