Granular fuzzy fractional descriptor linear systems under granular caputo fuzzy fractional derivative
Marzieh Najariyan,Yi Zhao
DOI: https://doi.org/10.1007/s00500-023-08549-3
IF: 3.732
2023-06-01
Soft Computing
Abstract:The objective of this research is to propose a fuzzy solution for a specific type of uncertain fractional descriptor linear systems referred to as granular fractional descriptor linear systems (gr-FFDLSs). The coefficients, boundary conditions, and parameters in gr-FFDLSs are represented as fuzzy numbers. This type of system also encompasses the concept of fractional order granular Caputo (gr-Caputo) fuzzy derivative. To achieve this goal, the gr-Caputo fuzzy derivative of order n-1<β<n\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$n-1<\beta <n$$\end{document}, where n is a natural number, is defined. Furthermore, the granular fuzzy Laplace transforms of the fractional order gr-Caputo fuzzy derivative and fractional order granular fuzzy integrals are introduced. Using the definition of fractional order gr-Caputo fuzzy derivative, fuzzy Drazin inverse matrix, and Laplace transformation, the solution to the state equation of granular descriptor fuzzy fractional linear systems is obtained. Two techniques, the fuzzy shuffle algorithm and the fuzzy Drazin inverse method, are presented to find the fuzzy solution of gr-FFDLSs. To demonstrate the effectiveness of these techniques, two examples are provided to show that they can produce accurate solutions for gr-FFDLSs.
computer science, artificial intelligence, interdisciplinary applications