Interval fuzzy modelling in fault detection for a class of processes with interval-type parameters

In the paper an application of the interval fuzzy model (INFUMO) in fault detection for a class of processes with uncertain interval-type parameters is presented. A confidence band for the process input-output data is obtained and approximated using a fuzzy model with interval parameters. The approximation is based on fuzzy linear programming using l_infin-norm as a measure of the modelling error. Applying high-pass filtering when obtaining the confidence band makes it possible to use arbitrary sets of identification input signals. An application of the INFUMO in detection of load change in a motor-generator system is presented to demonstrate the benefits of the proposed method