Diagnosis techniques for sensor faults of industrial processes

In this paper a model-based procedure exploitinganalytical redundancy for the detection and isolation of faults ininput--output control sensors of a dynamic system is presented.The diagnosis system is based on state estimators, namely dynamicobservers or Kalman filters designed in deterministic and stochasticenvironment respectively, and uses residual analysis and statisticaltests for fault detection and isolation.The state estimators are obtained from input--output data process andstandard identification techniques based on ARX orerrors--in--variables models, depending on signal to noise ratio. Inthe latter case the Kalman filter parameters, i.e. the modelparameters and the input--output noise variances, are obtained byprocessing the noisy data according to the Frisch scheme rules. The proposed fault detection and isolation tool has been tested on a single--shaft industrial gas turbine model. Results from simulation show that minimum detectable faults are perfectly compatible with the industrial target of this application.