Long-memory recursive prediction error method for identification of continuous-time fractional models

This paper deals with recursive continuous-time system identification using fractional-order models. Long-memory prediction error method is proposed for estimation of all parameters When differentiation orders are assumed known, least squares and methods, being direct extensions to models the classic methods used integer-order models, compared our new method, long-memory method. Given property fractional Monte Carlo simulations prove efficiency algorithm. Then, when unknown, two-stage algorithms necessary both parameter differentiation-order estimation. The performances algorithm studied through simulations. Finally, validated on a biological example where heat transfers in lungs modeled by thermal two-port network formalism