NN model-based evolved control by DGM model for practical nonlinear systems

• Published in: Expert systems with applications Vol. 193; p. 115873
• Main Authors: Chen, Z.Y., Meng, Yahui, Chen, Timothy
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.05.2022; Elsevier BV
• Subjects: Adaptive backstepping; Adaptive control; Control stability; Control systems design; Control theory; Controllers; Evolutionary algorithms; Evolved control; Feasibility; Fuzzy control; Fuzzy logic; Fuzzy systems; Gray predictor; Lyapunov energy function; Neural network; Neural networks; Nonlinear systems; Evolved control; Adaptive backstepping; Lyapunov energy function; Neural network; Gray predictor
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2021.115873

This article develops an evolutionary bat algorithm (EBA) based fuzzy neural network (NN), which is applied to the adaptive controller in a plant, for the purpose of ensuring the asymptotic stability and heightening the stability of the vehicle. With a gray signal predictor, the Lyapunov theory as well as the backstepping method are adopted to ensure the non-linearity of the controlled system and derive the evolutionary control law of the signal tracking. The discrete gray model (DGM) (2,1) is used to predict the future motion of the nonlinear system, so that the fuzzy controller can ensure the Lyapunov stability and the feasibility of the parallel distributed compensation (PDC) scheme through the Lyapunov type lemma. The controller design is demonstrated in practical nonlinear systems, for a mechanical elastic wheel (MEW), as a feasible mathematical framework for the matching of this circular object revolving on an axle.