Adaptive Self-Organizing Recurrent RBFN-Based Dynamic Surface Control for Linear Induction Motor Drive System with Dynamic Uncertainties

Abstract

In this paper, a robust adaptive dynamic surface control (RADSC) scheme is proposed to achieve high dynamic performance for linear induction motor (LIM) drives. The proposed control scheme comprises a dynamic surface controller (DSC), a self-organizing recurrent radial basis function network (SORRBFN) uncertainty estimator and a robust controller. First, an adaptive computed thrust controller (ACTC) is developed to stabilize the LIM drive system. However, the LIM drive performance may be degraded because all parameter uncertainties are not considered in the design of the ACTC. Therefore, the RADSC is proposed to improve the robustness of the LIM drive against all parameter uncertainties. In the RADSC, the DSC is used as the main tracking controller to overcome the explosion of the complexity in the backstepping design technique and the SORRBFN uncertainty estimator is designed to approximate the parameter uncertainties and compounded disturbances. In addition, the robust controller is designed to recover the approximation error of the SORRBFN. The online adaptive control laws are derived using the Lyapunov theory so that the stability of the closed-loop system is guaranteed. An experimental system is established and the control algorithms are implemented using a DSP-based control computer. The experimental results show the superiority of the proposed RADSC scheme in the presence of parameter uncertainties and compounded disturbances.