Robust fault detection based on compensation of the modelling error

The problem of compensation of modelling errors for the purpose of robust fault detection based on parity relations is addressed. The idea is to approximate unmodelled non-linear dynamics by a neural network model and then to remove the e€ ects of unmodelled dynamics from the primary residuals. The design of such a compensator takes two steps. In the ®rst, a subset of the most informative regressors is selected. The second step entails structure determination and parameter estimation by means of numerical optimization of a criterion function. The criterion re ̄ects a compromise between the quality of approximation and the complexity of the model structure. The results from the study on a three-tank test rig are presented and a comparison between compensated and uncompensated residuals made. It is shown that the compensated residuals represent a good basis for reliable, yet sensitive enough, fault detection and isolation.