Application of optimal adaptive control based on generalized minimum variance to a packed distillation column

Abstract

In this work, we present results from the successful application of optimal adaptive control with Generalized Minimum Variance method for packed distillation column in a pilot plant. The steady-state and dynamic behaviour of binary packed distillation column has been simulated using back mixing models. The model solution has been obtained employing orthogonal collocation on finite element methods. The solution region is considered to be consisting of small, interconnected subregions called finite elements and optimal oprating conditions were also determined by using steady-state model based on on-line optimization method. Optimal adaptive GMV control and PID methods were applied to the system. In all the control works, the heat input to the reboiler was taken as a manipulating variable. Generally, theoretical and experimental results were in agreement with each other and it was observed that GMV methods controlled the system better than PID control method.