A Hierarchical Identification Method for SISO Fractional-order State-space Systems

This paper proposes a new hierarchical identification method for fractional-order systems. In this method, a SISO (single input, single output) state space model has been considered in which parameters and also state variables should be estimated. By using a linear transformation and a shift operator, the system will be transformed into a form appropriate for identification of a fractional-order system. Then, the unknown parameters will be identified through a recursive least squares method and the states will be estimated using a fractional order Kalman filter. This identification method is based on the hierarchical identification principle that reduces the computational burden and is easy to implement on computer. The promising performance of the proposed method is verified using two stable fractional-order systems.