Stability Analysis of a Class of Stochastic Nonlinear Systems with Model Uncertainties Based on Neural Networks

Stochastic uncertainties and model uncertainties are often encountered and usually regarded as one of the most challenges in general practical system control problems. In this paper, we focus on the stability of a class of stochastic nonlinear systems with model uncertainties. A second-order disturbance observer is designed to compensate for system input time-varying disturbances. By using the second order of disturbance observer to measure fitting system input time-varying disturbances, using stochastic control Lyapunov method, the variance of uncertain Wiener noise system design the controller, eventually making system globally asymptotically stable. Simulation results verifies the effectiveness of the method and demonstrated a superior performance of the proposed control scheme in comparison to the other control scheme.