Neural network-based adaptive event-triggered sliding mode control for singular systems with an adaptive event-triggering communication scheme

• Published in: ISA transactions Vol. 129; no. Pt B; pp. 15 - 27
• Main Authors: Wang, Yuzhong, Zhang, Tie, Li, Jinna, Ren, Junchao
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.10.2022
• Subjects: Adaptive event-triggered SMC; Integral-type sliding surfaces; RBF neural network; Singular systems; Integral-type sliding surfaces; Adaptive event-triggered SMC; RBF neural network; Singular systems
• ISSN: 0019-0578; 1879-2022; 1879-2022
• DOI: 10.1016/j.isatra.2022.02.020

This paper studies the event-triggered sliding mode control problem for singular systems subject to the unknown nonlinear function and the exogenous disturbance. For saving the communication resources, a new adaptive event-triggering communication scheme (AETCS) is designed, which scheme uses the information on the nonlinear function part. Secondly, for the error system, we provide a novel integral sliding surface, which makes it beneficial to construct a new augmented delay system model by utilizing a delay system method. Furthermore, the sliding mode control (SMC) method for the error system is applied to compensate the unknown nonlinearity by using its estimate and match the exogenous disturbance by its upper bound. According to the Lyapunov function theory, stability criteria are got on the basis of LMIs. Moreover, two novel event-triggered adaptive sliding mode controllers based on RBF neural network are designed such that reachability conditions are obtained, and the asymptotic stability of singular systems with the H∞ performance is guaranteed. The RBF neural networks way is exploited to evaluate the unknown nonlinear function, which can eliminate the strict assumption of nonlinear function in some existing results. Finally, the proposed method is validated by two examples. •A novel AETCS based on the estimated weight of RBF NN is designed. This scheme uses the information on the nonlinear function part, and the connection between ETS and RBF NN is established for the first time.•Under the proposed AETCS, the SMC for the singular error system can compensate the unknown nonlinearity function by using its estimate and match the exogenous disturbance by its upper bound, and the effect of the estimation error can be neglected.•The adaptive method and the RBF NN method based on triggered signals are exploited to estimate the nonlinear f(x(t)) for singular system in case that assumptions of the Lipchitz conditions in Zhao et al. (2019)–Xia et al. (2019) can be removed.