#ICFDA23_GuelCortezetal

To date, the literature shows several methods in the time and frequency domain to estimate the parameters of a fractional-order model. One promising technique in the time domain corresponds to the fractional-order version of the so-called extended Kalman filter (EKF); Yet, the EKF is quite sensitive to the initial conditions of the state and error covariance matrices. A solution to this problem, when discussing integer order systems, consists of combining the EKF with evolutionary algorithms (EAs). The algorithm usually uses EAs to optimise the initial conditions for the EKF and leads to a better estimate of the system parameters and states. Here, we show how this methodology can be extended to fractional-order models. Specifically, we create a scheme combining a GA and EKF to estimate the values of model parameters, including the fractional order, and show its application to a simple mechanical model.

Fractional-order Systems, Parameter Identification, Extended Kalman Filter, Genetic Algorithms, Fractional Calculus