Fault detection and diagnosis for the non-Gaussian time-delayed stochastic distribution control system

Stochastic distribution control is a new branch of stochastic system control. The main feature of this system is that the system output is the output probability density function. The control goal of such stochastic distribution systems is to choose an input to make the output probability density function track a given probability density function. In practice, time delay may exist in the modeling, the data (or image) processing and transmission phases. The effectiveness of the fault detection and diagnosis will be reduced when time delay is not considered. In this paper, the rational square-root B-spline is used to approach the output probability density function. A fault diagnosis method based on the nonlinear adaptive observer is used to diagnose the fault of the nonlinear stochastic distribution system with state delay. Finally computer simulations are given to demonstrate the effectiveness of the proposed algorithm.