Dead Time Compensation via the Time Domain Predictor Method

In this paper a time domain predictor is introduced and identified as a reduction transformation. Properties of the time domain predictor method for dead time compensation are investigated and compared with the well known Smith predictor method. An output feedback control law is also suggested by intoducing an observer for the predictor system. It has been shown in this paper that above methods can be applied to general delay systems with multiple delays in control as well as in measurement. The methods suggested in this paper have several advantages over the well known Smith predictor methods.