Stable inverse control for nonminimum-phase nonlinear processes

Since most of the chemical processes happen to be nonlinear, nonlinear controllers with feedback linearization have been recently designed which can directly cancel the nonlinearities and establish a linear input/output map. However, this I/O linearization algorithm requires the control with stable inversion and state feedback. In the present study, a stable inversion technique for nonminimum-phase systems is developed under appropriate assumptions, which includes several steps such as finding the bounded control and bounded state trajectories for tracking the desired signal. In this design, the output tracking error as well as the effect of measurable disturbances can be asymptotically eliminated through generating a feedforward control. Finally, the proposed control methodology is applied to a chemical process, viz. a Van de Vusse reaction in an isothermal or adiabatic stirred-tank reactor, and its performance is evaluated.