Interaction among control loops in multivariable systems has been the subject of much research in the last decades. Distillation columns that are widely used in the process industries are the typical example of these systems. In this work, Wood and Berry test-bed problem has been used due to the strong interaction among the control variables. Therefore, different control strategies have been applied to this binary distillation column with two interacting control loops. A diagonal controller structure with proportional-integral controllers using several tuning methods has been evaluated. Two Internal Model Control structures have been implemented utilizing in one of them a neural network model of the plant. Model identification by neural networks has also been presented.