CONICET | Buscador de Institutos y Recursos Humanos

In this work, a plant-wide control methodology, named Minimum Square Deviation (MSD), is applied on the challenging Petlyuk distillation column. This process presents an interesting design since allows to separate a ternary mixture in the same shell, reducing up to 30% of the energy consumption and the capital investment cost respect to classic layout of two integrated columns. However, strong dynamic control problems are reported, caused by the intrinsic coupling between the closed loops proposed to achieve the control objectives. The MSD method is able to select systematically the controlled variables and the controller structure via the sum of square deviations (SSD) and net load evaluation (NLE) indexes respectively. From SSD procedure a diagonal control structure is achieved. Meanwhile, the NLE stage is a valuable tool for deciding which is the best control configuration among full, diagonal or sparse, according to the relative importance between servo and regulator problems. On the other hand, self optimizing control (SOC), a well recognized plantwide control strategy, was implemented previously on the Petlyuk distillation column. Different controlled variables were selected by applying singular value and null space methods which gave different decentralized control structures. Thus, one of the contributions of this work is to give a new alternative to control the Petlyuk distillation column by using MSD approach. The second contribution is to quantify the skills of both methods, MSD and SOC, through the use of the same complex case. The third contribution is to propose some improvements on plant-wide control techniques by combining both methodologies, MSD and SOC. Hence, a complete set of simulation and quantitative comparisons are given for several control structures and scenarios to give support of the final conclusions.