Application of a neural network in gas turbine control sensor fault detection

In this paper an application of a procedure using a neural network for the detection and isolation of faults modeled by step functions in input-output control sensors of a single shaft industrial gas turbine is presented. The real process is modeled as a linear dynamic system corrupted by stochastic additive noise. The diagnosis system involves dynamic observers and utilizes the neural network in order to classify observer residuals into fault classes