The Design of Adaptive Sliding Mode Controller Based on RBFNN Approximation for Suspension Control of MVAWT

The maglev system (MS) is one of the key parts of the maglev vertical axis wind turbine (MVAWT). An adaptive sliding mode controller based on RBF neural network approximation (RBFNN-ASMC) is proposed to enhance the dynamic performance and anti-interference ability of the MS. First, the mathematical model of the MS is given. Second, because the MS is easily affected by wind loads such as gusts, crosswind etc., RBFNN is employed to approximate these unknown random disturbances. Meanwhile, the bipolar sigmoid function is proposed to substitute for the sign function in the exponential approach law of the traditional SMC, so that the sliding mode surface can change more smoothly. Moreover, the Lyapunov stability theorem is adopted to design the adaptive law of the MS. Finally, the simulations are executed and the results indicate that the control strategy has faster dynamic response and better robustness compared with the traditional dual-loop PID (DL-PID) and the dual-loop sliding mode control (DL-SMC). The proposed controller can satisfy the requirement for the steady suspension operation of the MVAWT.