USP Electronic Research Repository

Modeling and parametric identification of Hammerstein systems with time delay and asymmetric dead - zones using fractional differential equations

Prasad, Vineet and Mehta, Utkal V. (2022) Modeling and parametric identification of Hammerstein systems with time delay and asymmetric dead - zones using fractional differential equations. Mechanical Systems and Signal Processing, 167 . NA. ISSN 0888-3270

[img] PDF - Published Version
Restricted to Registered users only

Download (1638Kb)

    Abstract

    The parametric identification of Hammerstein structured nonlinear systems with discontinuous asymmetric (two segment piecewise-linear with a dead-zone) nonlinearity and input time delay is presented using a fractional-order modeling technique. The effect of the unknown dead-zone nonlinearity is separated from the linear dynamics using special excitation signals to simplify the identification process. The use of fractional calculus permits reduced order modeling of the linear dynamics. An interactive block-based strategy is then introduced to simultaneously estimate the separated nonlinear and linear parameters, including the fractional differentiation order(s) and input time delay, which usually require separate algorithms in such cases. The proposed method utilizes block pulse functions to mitigate the computational complexity of fractional differential operations. Numerical examples are provided to validate the efficacy in comparison to existing methods. A DC servo motor application is used to demonstrate the proposed nonlinear modeling approach.

    Item Type: Journal Article
    Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > Robotics and Automation
    Divisions: School of Information Technology, Engineering, Mathematics and Physics (STEMP)
    Depositing User: Utkal Mehta
    Date Deposited: 29 Nov 2021 15:58
    Last Modified: 29 Nov 2021 15:58
    URI: http://repository.usp.ac.fj/id/eprint/13119
    UNSPECIFIED

    Actions (login required)

    View Item

    Document Downloads

    More statistics for this item...