Neuroadaptive dynamic surface control for induction motors stochastic system based on reduced-order observer

This paper studies an observer-based neural network position tracking control scheme for induction motors system operating under a field-oriented control scheme with the problem of stochastic disturbance. Firstly, the angular velocity is estimated by the constructed reduced-order observer. Then, the nonlinear functions are approximated by the neural networks and the stochastic Lyapunov functions are chosen to analyze the stability of the system. Besides, the "complexity of computation" existed in traditional backstepping control is solved by using the dynamic surface control technique. At last, the results of the comparison simulation experiments show that the proposed control scheme can reduce the influence of stochastic disturbance, and have faster tracking speed smaller tracking error. The designed observer can estimate the signals effectively.