Structural Identifiability Analysis of Fractional Order Models with Applications in Battery Systems
Abstract
This paper presents a method for structural identifiability analysis of fractional order systems by using the coefficient mapping concept to determine whether the model parameters can uniquely be identified from input-output data. The proposed method is applicable to general non-commensurate fractional order models. Examples are chosen from battery fractional order equivalent circuit models (FO-ECMs). The battery FO-ECM consists of a series of parallel resistors and constant phase elements (CPEs) with fractional derivatives appearing in the CPEs. The FO-ECM is non-commensurate if more than one CPE is considered in the model. Currently, estimation of battery FO-ECMs is performed mainly by fitting in the frequency domain, requiring costly electrochemical impedance spectroscopy equipment. This paper aims to analyse the structural identifiability of battery FO-ECMs directly in the time domain. It is shown that FO-ECMs with finite numbers of CPEs are structurally identifiable. In particular, the FO-ECM with a single CPE is structurally globally identifiable.
Keywords
Cite
@article{arxiv.1511.01402,
title = {Structural Identifiability Analysis of Fractional Order Models with Applications in Battery Systems},
author = {S. M. Mahdi Alavi and Adam Mahdi and Pierre E. Jacob and Stephen J. Payne and David A. Howey},
journal= {arXiv preprint arXiv:1511.01402},
year = {2015}
}