Authors: Cuiying Yan; Yang Li; Jianhua Zhang
Addresses: Hebei University of Science and Technology, Shijiazhuang, 050054, China ' YanShan University, Qinhuangdao, 066004, China ' Hebei University of Science and Technology, Shijiazhuang, 050054, China
Abstract: This study develops a generalised procedure in adaptive neural network enhanced controller design for strict feedback non-linear time delay systems. Under the framework, recurrent neural network is tailored to accommodate the on-line identification, by which the weights of the neural network are iteratively and adaptively updated through the system state. Based on the neural network online approximation model, the state feedback adaptive controller is obtained by constructing a Lyapunov-Krasovskii function, which the integral type efficiently overcomes the controller singularity problem. To guarantee the correctness, rigorousness, and generality of the developed results, Lyapunov stability theory is used to prove the closed-loop control systems semiglobally, uniformly, and ultimately bounded stable. Two bench mark tests are simulated to demonstrate the effectiveness and efficiency of the procedure and furthermore these could be the showcases for potential users to apply to their demanded tasks.
Keywords: neural networks; time delay systems; strict feedback; nonlinear systems; adaptive control; backstepping; simulation.
International Journal of Modelling, Identification and Control, 2014 Vol.21 No.4, pp.401 - 410
Available online: 29 May 2014 *Full-text access for editors Access for subscribers Purchase this article Comment on this article