Recurrent general type-2 fuzzy neural networks for nonlinear dynamic systems identification

Ahmad M El-Nagar; Mohammad El-Bardini; A Aziz Khater

doi:10.1016/j.isatra.2023.06.003

Recurrent general type-2 fuzzy neural networks for nonlinear dynamic systems identification

ISA Trans. 2023 Sep:140:170-182. doi: 10.1016/j.isatra.2023.06.003. Epub 2023 Jun 5.

Authors

Ahmad M El-Nagar¹, Mohammad El-Bardini², A Aziz Khater³

Affiliations

¹ Department of Industrial Electronics and Control Engineering, Faculty of Electronic Engineering, Menofia University, Menof, 32852, Egypt. Electronic address: ahmed.elnagar@el-eng.menofia.edu.eg.
² Department of Industrial Electronics and Control Engineering, Faculty of Electronic Engineering, Menofia University, Menof, 32852, Egypt. Electronic address: dralbardini@el-eng.menofia.edu.eg.
³ Department of Industrial Electronics and Control Engineering, Faculty of Electronic Engineering, Menofia University, Menof, 32852, Egypt. Electronic address: abdelaziz.khater@el-eng.menofia.edu.eg.

PMID: 37328315
DOI: 10.1016/j.isatra.2023.06.003

Abstract

This paper introduces a recurrent general type-2 Takagi-Sugeno-Kang fuzzy neural network (RGT2-TSKFNN) for the identification of nonlinear systems. In the proposed structure, the general type-2 fuzzy set (GT2FS) and a recurrent fuzzy neural network (RFNN) are combined to obviate the data uncertainties. The fuzzy firing strengths in the developed structure are returned to the network input as internal variables. In the proposed structure, GT2FS is utilized to characterize the antecedent parts while the consequent parts are performed using TSK type. The issues of constructing a RGT2-TSKFNN involve type reduction, structure learning as well as parameter learning. An efficient strategy is developed by utilizing alpha-cuts to decompose a GT2FS into several interval type-2 fuzzy sets (IT2FSs). In order to solve the computation time of the type-reduction issue, a direct defuzzification method is used instead of iterative nature of Karnik-Mendel (KM) algorithm. Type-2 fuzzy clustering and Lyapunov criteria are utilized for online structure learning as well as the antecedent and consequent parameters, respectively for reducing the number of rules and guaranteeing the stability of the proposed RGT2-TSKFNN. The reported comparative analysis of the simulation results is utilized to estimate the performance of the proposed RGT2-TSKFNN with respect to other popular type-2 FNNs (T2FNNs) methodologies.

Keywords: Alpha planes; Fuzzy neural networks; General type-2 fuzzy sets; Lyapunov function; Online learning; System identification.