Abstract
For a class of continuous time plant with uncertainties, a new control system that integrates plant identification and feedback control by using neural network is presented in this paper. In the proposed control system, an identification model based on a neural network (NN) is connected in parallel with a plant model. The NN can identify the plant uncertainties and can modify a control signal to the plant for overcoming effect of the uncertainties simultaneously. By the Lyapunov stability techniques, stability analysis of the proposed control system is shown and a sufficient condition of the asymptotical stability is derived in case of a linear plant. Computer simulations are performed to illustrate the effectiveness and applicability of the proposed control system to a variety of continuous time plants with linear and nonlinear uncertainties.
