A New FDD Algorithm of a Class of Nonlinear Non-Gaussian Stochastic Systems

A new fault detection and diagnosis (FDD) algorithm for general nonlinear stochastic systems is proposed by using the optimal probability density function (PDF) tracking filtering. The fault is detected through a determinate threshold rather than an experiential threshold. Moreover, an adaptive fault diagnosis method is also provided to estimate the size of the fault. Specially, to give facilities for practical application, a time-varying threshold (TVT), which can be determined beforehand, is presented. Simulations are included to show the effectiveness of the proposed algorithm under the missing measurements and encouraging results have been obtained via comparison to existing detection algorithms.