Systematic Optimization Methodology for Simultaneous Process Design and Heat Exchanger Network Synthesis with Surrogate Models – Application to Distillation Column Complex

A. Pagot,Andreas Psaltis,D. Sinoquet
Abstract:Distillation units require huge amounts of energy for the separation of the multicomponent mixtures involved in refineries and petrochemical industries. Keeping the same separation standards, energy savings can be achieved in multiple ways. The overall efficiency of the distillation column system is determined from the trade-offs of the Operating Expenditures (OPEX) and Capital investment cost (CAPEX), as there is a strong interaction between the distillation columns and the Heat Exchanger Network (HEN) of the interconnecting streams. Therefore, the highly complex optimization of this trade-off necessitates moving towards process integration solutions. Individually, both distillation columns and HEN are well-defined rigorous problems and numerous methodologies have been developed for finding optimal solutions. The integrative problem however is larger in scale, computational demanding and less robust. In this paper, a systematic optimization methodology for process integration of a multicomponent distillation column complex is presented. The highly nonlinear and time consuming rigorous models of the distillation column are being substituted with simple surrogate models that generate operating responses with adequate accuracy. Moreover, another surrogate model is included in the Mixed Integer Non-Linear Programming (MINLP) formulation to enhance the accuracy of the results by considering the phase changes of the process streams involved in the HEN. The methodology is applied on two case studies of the aromatics separation PARAMAX complex and the results illustrate significant reductions on the Total Annualized Cost. With a scope limited to the benzene and toluene columns, the gain reaches about 15%.
Engineering,Environmental Science,Chemistry
What problem does this paper attempt to address?