Robust Virtual Sensors Design

The problem of robust virtual sensors design in linear systems under disturbance is studied. The suggested approach based on the reduced order model of the original system and singular value decomposition enables obtaining virtual sensor of minimal dimension insensitive or having minimal sensitivity to the disturbance. The problems is solved in three steps: at the first step, the linear model insensitive to the disturbance is designed; at the second step, a possibility to estimate the given variable is checked; finally, sliding mode observer is constructed. If the model insensitive to the disturbance does not exist, the minimal sensitivity is achieved. The theoretical results are illustrated by the example of three-tank system.