Title :
Identification of fractional-order system based on modified differential evolution
Author :
Lilian Huang ; Xiaoliang Zhou
Author_Institution :
Coll. of Inf. & Commun. Eng., Harbin Eng. Univ., Harbin, China
Abstract :
Fractional order system can accurately describe systems by introducing additional degrees of freedom, but at the same time it increases the difficulty to identify the systems. In this paper, a modified differential evolution(MDE) is proposed, according to the response data of time domain, the rational fitness function is designed to identify the fractional order SISO(Single Input Single Output) system. The proposed method could identify both the parameters and the orders of the model, and is adaptive to the unknown system models. The simulation results show the accuracy and efficiency of this method.
Keywords :
evolutionary computation; parameter estimation; time-domain analysis; MDE; fractional order SISO identification; fractional order single input single output system; modified differential evolution; parameter identification; rational fitness function; time domain response data; Convergence; Data models; Mathematical model; Sociology; Statistics; Time-domain analysis; Vectors; Fitness Function; Fractional-order System; Modified Differential Evolution; SISO(Single Input Single Output); Time Domain Identification;
Conference_Titel :
Control and Decision Conference (CCDC), 2013 25th Chinese
Conference_Location :
Guiyang
Print_ISBN :
978-1-4673-5533-9
DOI :
10.1109/CCDC.2013.6561235