Dynamic self-adaptive learning algorithm research based on T-S RBF fuzzy neutral network

Article type: Research Article

Authors: Pu, Cuiping^{a; *} | Ren, Jie^a | Xue, Bin^b | Yu, An^a

Affiliations: [a] Kunming University, Kunming, Yunnan, China | [b] Phicomm (Shanghai) Co., Ltd, Shanghai, China

Correspondence: [*] Corresponding author: Cuiping Pu, Kunming University, Kunming, Yunnan, China. E-mail: [email protected].

Abstract: The paper makes deep research about T-S fuzzy model, BP neural network and RBF neural network respectively first. After simulation experiments, RBF network has more advantages than BP network in nonlinear system identification and control. Then, combing T-S fuzzy model with RBF network organizationally, RBF fuzzy neural network based on T-S fuzzy model is gotten and a kind of dynamic study arithmetic is put forward at the same time. Actual simulation shows it’s able to approach arbitrary nonlinear object and construct model which has good control effects and strong fault-tolerant and robustness.

Keywords: T-S model fuzzy inference system, RBF network, predict control

DOI: 10.3233/JCM-180883

Journal: Journal of Computational Methods in Sciences and Engineering, vol. 19, no. 1, pp. 209-218, 2019