Study On The Optimal Operation Of Dividing-wall Column Via Steady-state Analysis

Dividing-wall column commonly offers significant energy and capital savings(typically around30%) compared with conventional two-column sequence forseparating some ternary mixtures. Previous studies have focused on ensuring productpurity in operation, and neglected the full play of energy-saving potential of DWC. Inthis paper, the full play of energy-saving capabilities and purity guaranteed are takeninto account simultaneously. The low-energy region of DWC is traced by steady-stateanalysis, and this will guide us to achieve the optimal operation of DWC.In this paper, a method of steady-state analysis of the DWC based on rigorousmodel is presented. To illustrate the efficacy of our method, we apply it to theindustrially important ternary separation of benzene, toluene, and o-xylene. Firstly,the rigorous model of DWC is built, and the relation between the variables of DWCsystem is obtained. Then based on the relation between the actual energy consumptionand minimum energy consumption, the low-energy region of DWC is quantified.Three manifestations of the low-energy region are proposed. Using the numericalmethod to solve the optimization model of DWC, the low-energy region is get. Theselection range of manipulated variable for energy saving is determined, throughanalyzing the steady-state behavior of DWC under the base-case feed condition. Thesensitivity analysis of low-energy region with feed condition show that the DWC needon-line optimization to realize the optimal operation. And the theoretical exertingdegree of energy-saving potential for single-variables-optimization operation mode isobtained from the sensitivity analysis. The feedback method is used for optimizing theDWC. The concentration of light component at the bottom of prefractionator and theconcentration of heavy component at the top of the prefractionator are selected ascontrolled variables and the reason for the selection is given. The setpoint of the othermanipulated variable, the setpoint of controlled variable and exerting degree ofenergy-saving potential for single-variable-optimization operation mode are obtainedby the sensitivity analysis of hybrid low-energy region with feed conditions. Thesetpoints of controlled variables and exerting degree of energy-saving potential fordouble-variables-optimization operation mode are obtained by the sensitivity analysisof setpoint low-energy region with feed conditions.