A REDUCED ORDER MODEL SIMULATION FOR DISTILLATION COLUMN

Abstract

The simulation of dynamic model of a distillation column is required for design of control schemes, online identification and optimization purposes. The full order stagewise model requires the solution of mixed non-linear programming problem. The problem is compounded by the fact that the number of equations varies with the number of stages. The solution of such a system of differential equation is complicated because algebraic equations being non-linear and stiff. The reduced order model is used in order to reduce the total number of equations. In present work reduce order model is developed for distillation column to reduce the dimensionality of the problem, so it can be used for online identification, optimization and control with little loss in accuracy compared with full order model and also predicts the state variable at any stage. The reduce order model for distillation column is developed with the method of orthogonal collocation method. In which Jacobi polynomial is used to derive location of collocation points and Lagrange's interpolation method for approximating the variables at each collocation point. Further the developed reduced order model is tested for a distillation column having 30 trays with a five component mixture. The results are compared with that of full order model.