Modeling and identification of fractional-order dynamical systems

• Main Authors: Dorčák, Ľubomír, Technical University of Kosice, Valsa, Juraj, Terpák, Ján, Horovčák, Pavel, Gonzalez, Emmanuel
• Format: text
• Published: Animo Repository 2011
• Subjects: Fractional differential equations; Laplace transformation; Computer Sciences; Numerical Analysis and Scientific Computing
• Online Access: https://animorepository.dlsu.edu.ph/faculty_research/3179
• Institution: De La Salle University

One of the most characteristic features of the processes in the area of mining and processing of earth resources is their complexity. The modern methods of investigation, monitoring and control of these processes utilize approaches based upon modeling of the processes. In this paper we will focus only on the dynamical processes, where the mathematical models are differential equations. As it results from the last research works, majority of the real objects in general are fractional-order systems. Fractional order models are more adequate for the description of dynamical systems than integer order models. But appropriate methods for the numerical calculations of fractional-order differential equations are needed in such cases and also methods for the identification of such systems. In this contribution we describe and compare some previous methods used for simulation purposes with the methods based on approximate formulas for numerical inversion of Laplace transforms. The verification and comparison will be based mainly on the accuracy and computing time. Also an experimental method of identification is given. The verification of identification method was performed on systems with known parameters. © SGEM2011 All Rights Reserved by the International Multidisciplinary Scientific GeoConference SGEM.